program(1.3) [buildInfo = dict({{"coremlc-component-MIL", "3600.16.1"}, {"coremlc-version", "3600.25.1"}, {"coremltools-component-torch", "2.8.0"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "9.0"}})] { func main(tensor input_features, tensor key_bias) { tensor var_63 = const()[name = string("op_63"), val = tensor([1, 128, 8, 100])]; tensor var_64_cast_fp16 = reshape(shape = var_63, x = input_features)[name = string("op_64_cast_fp16")]; tensor var_69 = const()[name = string("op_69"), val = tensor([0, 2, 1, 3])]; tensor var_75 = const()[name = string("op_75"), val = tensor([8, 1, 128, 100])]; tensor var_70_cast_fp16 = transpose(perm = var_69, x = var_64_cast_fp16)[name = string("transpose_4226")]; tensor input_1_cast_fp16 = reshape(shape = var_75, x = var_70_cast_fp16)[name = string("input_1_cast_fp16")]; string x_1_pad_type_0 = const()[name = string("x_1_pad_type_0"), val = string("custom")]; tensor x_1_pad_0 = const()[name = string("x_1_pad_0"), val = tensor([1, 1, 1, 1])]; tensor x_1_strides_0 = const()[name = string("x_1_strides_0"), val = tensor([2, 2])]; tensor x_1_dilations_0 = const()[name = string("x_1_dilations_0"), val = tensor([1, 1])]; int32 x_1_groups_0 = const()[name = string("x_1_groups_0"), val = int32(1)]; tensor conv2d1_weight_to_fp16 = const()[name = string("conv2d1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(64)))]; tensor conv2d1_bias_to_fp16 = const()[name = string("conv2d1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(8768)))]; tensor x_1_cast_fp16 = conv(bias = conv2d1_bias_to_fp16, dilations = x_1_dilations_0, groups = x_1_groups_0, pad = x_1_pad_0, pad_type = x_1_pad_type_0, strides = x_1_strides_0, weight = conv2d1_weight_to_fp16, x = input_1_cast_fp16)[name = string("x_1_cast_fp16")]; fp16 var_89_to_fp16 = const()[name = string("op_89_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_90_cast_fp16 = mul(x = x_1_cast_fp16, y = var_89_to_fp16)[name = string("op_90_cast_fp16")]; tensor var_91_cast_fp16 = mul(x = var_90_cast_fp16, y = x_1_cast_fp16)[name = string("op_91_cast_fp16")]; tensor var_92_cast_fp16 = mul(x = var_91_cast_fp16, y = x_1_cast_fp16)[name = string("op_92_cast_fp16")]; tensor var_94_cast_fp16 = add(x = x_1_cast_fp16, y = var_92_cast_fp16)[name = string("op_94_cast_fp16")]; fp16 var_95_to_fp16 = const()[name = string("op_95_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_1_cast_fp16 = mul(x = var_94_cast_fp16, y = var_95_to_fp16)[name = string("u_1_cast_fp16")]; fp16 var_97_to_fp16 = const()[name = string("op_97_to_fp16"), val = fp16(0x1p-1)]; tensor var_98_cast_fp16 = mul(x = x_1_cast_fp16, y = var_97_to_fp16)[name = string("op_98_cast_fp16")]; tensor var_99_cast_fp16 = tanh(x = u_1_cast_fp16)[name = string("op_99_cast_fp16")]; fp16 var_101_to_fp16 = const()[name = string("op_101_to_fp16"), val = fp16(0x1p+0)]; tensor var_102_cast_fp16 = add(x = var_99_cast_fp16, y = var_101_to_fp16)[name = string("op_102_cast_fp16")]; tensor input_3_cast_fp16 = mul(x = var_98_cast_fp16, y = var_102_cast_fp16)[name = string("input_3_cast_fp16")]; string x_3_pad_type_0 = const()[name = string("x_3_pad_type_0"), val = string("custom")]; tensor x_3_pad_0 = const()[name = string("x_3_pad_0"), val = tensor([1, 1, 1, 1])]; tensor x_3_strides_0 = const()[name = string("x_3_strides_0"), val = tensor([2, 2])]; tensor x_3_dilations_0 = const()[name = string("x_3_dilations_0"), val = tensor([1, 1])]; int32 x_3_groups_0 = const()[name = string("x_3_groups_0"), val = int32(1)]; tensor conv2d2_weight_to_fp16 = const()[name = string("conv2d2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(9792)))]; tensor conv2d2_bias_to_fp16 = const()[name = string("conv2d2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(4157056)))]; tensor x_3_cast_fp16 = conv(bias = conv2d2_bias_to_fp16, dilations = x_3_dilations_0, groups = x_3_groups_0, pad = x_3_pad_0, pad_type = x_3_pad_type_0, strides = x_3_strides_0, weight = conv2d2_weight_to_fp16, x = input_3_cast_fp16)[name = string("x_3_cast_fp16")]; fp16 var_116_to_fp16 = const()[name = string("op_116_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_117_cast_fp16 = mul(x = x_3_cast_fp16, y = var_116_to_fp16)[name = string("op_117_cast_fp16")]; tensor var_118_cast_fp16 = mul(x = var_117_cast_fp16, y = x_3_cast_fp16)[name = string("op_118_cast_fp16")]; tensor var_119_cast_fp16 = mul(x = var_118_cast_fp16, y = x_3_cast_fp16)[name = string("op_119_cast_fp16")]; tensor var_121_cast_fp16 = add(x = x_3_cast_fp16, y = var_119_cast_fp16)[name = string("op_121_cast_fp16")]; fp16 var_122_to_fp16 = const()[name = string("op_122_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_3_cast_fp16 = mul(x = var_121_cast_fp16, y = var_122_to_fp16)[name = string("u_3_cast_fp16")]; fp16 var_124_to_fp16 = const()[name = string("op_124_to_fp16"), val = fp16(0x1p-1)]; tensor var_125_cast_fp16 = mul(x = x_3_cast_fp16, y = var_124_to_fp16)[name = string("op_125_cast_fp16")]; tensor var_126_cast_fp16 = tanh(x = u_3_cast_fp16)[name = string("op_126_cast_fp16")]; fp16 var_128_to_fp16 = const()[name = string("op_128_to_fp16"), val = fp16(0x1p+0)]; tensor var_129_cast_fp16 = add(x = var_126_cast_fp16, y = var_128_to_fp16)[name = string("op_129_cast_fp16")]; tensor input_5_cast_fp16 = mul(x = var_125_cast_fp16, y = var_129_cast_fp16)[name = string("input_5_cast_fp16")]; string x_5_pad_type_0 = const()[name = string("x_5_pad_type_0"), val = string("custom")]; tensor x_5_pad_0 = const()[name = string("x_5_pad_0"), val = tensor([1, 1, 1, 1])]; tensor x_5_strides_0 = const()[name = string("x_5_strides_0"), val = tensor([2, 2])]; tensor x_5_dilations_0 = const()[name = string("x_5_dilations_0"), val = tensor([1, 1])]; int32 x_5_groups_0 = const()[name = string("x_5_groups_0"), val = int32(1)]; tensor conv2d3_weight_to_fp16 = const()[name = string("conv2d3_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(4158080)))]; tensor conv2d3_bias_to_fp16 = const()[name = string("conv2d3_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(8305344)))]; tensor x_5_cast_fp16 = conv(bias = conv2d3_bias_to_fp16, dilations = x_5_dilations_0, groups = x_5_groups_0, pad = x_5_pad_0, pad_type = x_5_pad_type_0, strides = x_5_strides_0, weight = conv2d3_weight_to_fp16, x = input_5_cast_fp16)[name = string("x_5_cast_fp16")]; fp16 var_143_to_fp16 = const()[name = string("op_143_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_144_cast_fp16 = mul(x = x_5_cast_fp16, y = var_143_to_fp16)[name = string("op_144_cast_fp16")]; tensor var_145_cast_fp16 = mul(x = var_144_cast_fp16, y = x_5_cast_fp16)[name = string("op_145_cast_fp16")]; tensor var_146_cast_fp16 = mul(x = var_145_cast_fp16, y = x_5_cast_fp16)[name = string("op_146_cast_fp16")]; tensor var_148_cast_fp16 = add(x = x_5_cast_fp16, y = var_146_cast_fp16)[name = string("op_148_cast_fp16")]; fp16 var_149_to_fp16 = const()[name = string("op_149_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_5_cast_fp16 = mul(x = var_148_cast_fp16, y = var_149_to_fp16)[name = string("u_5_cast_fp16")]; fp16 var_151_to_fp16 = const()[name = string("op_151_to_fp16"), val = fp16(0x1p-1)]; tensor var_152_cast_fp16 = mul(x = x_5_cast_fp16, y = var_151_to_fp16)[name = string("op_152_cast_fp16")]; tensor var_153_cast_fp16 = tanh(x = u_5_cast_fp16)[name = string("op_153_cast_fp16")]; fp16 var_155_to_fp16 = const()[name = string("op_155_to_fp16"), val = fp16(0x1p+0)]; tensor var_156_cast_fp16 = add(x = var_153_cast_fp16, y = var_155_to_fp16)[name = string("op_156_cast_fp16")]; tensor x_7_cast_fp16 = mul(x = var_152_cast_fp16, y = var_156_cast_fp16)[name = string("x_7_cast_fp16")]; tensor var_162 = const()[name = string("op_162"), val = tensor([8, 7680, 1, 13])]; tensor input_7_cast_fp16 = reshape(shape = var_162, x = x_7_cast_fp16)[name = string("input_7_cast_fp16")]; string x_9_pad_type_0 = const()[name = string("x_9_pad_type_0"), val = string("valid")]; tensor x_9_strides_0 = const()[name = string("x_9_strides_0"), val = tensor([1, 1])]; tensor x_9_pad_0 = const()[name = string("x_9_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_9_dilations_0 = const()[name = string("x_9_dilations_0"), val = tensor([1, 1])]; int32 x_9_groups_0 = const()[name = string("x_9_groups_0"), val = int32(1)]; tensor conv_out_weight_to_fp16 = const()[name = string("conv_out_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(8306368)))]; tensor x_9_cast_fp16 = conv(dilations = x_9_dilations_0, groups = x_9_groups_0, pad = x_9_pad_0, pad_type = x_9_pad_type_0, strides = x_9_strides_0, weight = conv_out_weight_to_fp16, x = input_7_cast_fp16)[name = string("x_9_cast_fp16")]; tensor pos_ct_to_fp16 = const()[name = string("pos_ct_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(24035072)))]; tensor x_11_cast_fp16 = add(x = x_9_cast_fp16, y = pos_ct_to_fp16)[name = string("x_11_cast_fp16")]; tensor var_181 = const()[name = string("op_181"), val = tensor([1, 2, 0, 3])]; tensor var_187 = const()[name = string("op_187"), val = tensor([1, 1024, 1, 104])]; tensor var_182_cast_fp16 = transpose(perm = var_181, x = x_11_cast_fp16)[name = string("transpose_4225")]; tensor x_13_cast_fp16 = reshape(shape = var_187, x = var_182_cast_fp16)[name = string("x_13_cast_fp16")]; int32 var_195 = const()[name = string("op_195"), val = int32(1)]; tensor mu_1_axes_0 = const()[name = string("mu_1_axes_0"), val = tensor([1])]; bool mu_1_keep_dims_0 = const()[name = string("mu_1_keep_dims_0"), val = bool(true)]; tensor mu_1_cast_fp16 = reduce_mean(axes = mu_1_axes_0, keep_dims = mu_1_keep_dims_0, x = x_13_cast_fp16)[name = string("mu_1_cast_fp16")]; tensor var_209_cast_fp16 = sub(x = x_13_cast_fp16, y = mu_1_cast_fp16)[name = string("op_209_cast_fp16")]; fp16 var_198_promoted_to_fp16 = const()[name = string("op_198_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_210_cast_fp16 = pow(x = var_209_cast_fp16, y = var_198_promoted_to_fp16)[name = string("op_210_cast_fp16")]; tensor var_1_axes_0 = const()[name = string("var_1_axes_0"), val = tensor([1])]; bool var_1_keep_dims_0 = const()[name = string("var_1_keep_dims_0"), val = bool(true)]; tensor var_1_cast_fp16 = reduce_mean(axes = var_1_axes_0, keep_dims = var_1_keep_dims_0, x = var_210_cast_fp16)[name = string("var_1_cast_fp16")]; fp16 var_214_to_fp16 = const()[name = string("op_214_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_215_cast_fp16 = add(x = var_1_cast_fp16, y = var_214_to_fp16)[name = string("op_215_cast_fp16")]; fp32 var_216_epsilon_0 = const()[name = string("op_216_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_216_cast_fp16 = rsqrt(epsilon = var_216_epsilon_0, x = var_215_cast_fp16)[name = string("op_216_cast_fp16")]; tensor x_15_cast_fp16 = mul(x = var_209_cast_fp16, y = var_216_cast_fp16)[name = string("x_15_cast_fp16")]; tensor input_9_mean_0_to_fp16 = const()[name = string("input_9_mean_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(24061760)))]; tensor input_9_variance_0_to_fp16 = const()[name = string("input_9_variance_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(24063872)))]; tensor input_9_gamma_0_to_fp16 = const()[name = string("input_9_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(24065984)))]; tensor input_9_beta_0_to_fp16 = const()[name = string("input_9_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(24068096)))]; fp16 input_9_epsilon_0_to_fp16 = const()[name = string("input_9_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_9_cast_fp16 = batch_norm(beta = input_9_beta_0_to_fp16, epsilon = input_9_epsilon_0_to_fp16, gamma = input_9_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_15_cast_fp16)[name = string("input_9_cast_fp16")]; string var_234_pad_type_0 = const()[name = string("op_234_pad_type_0"), val = string("valid")]; tensor var_234_strides_0 = const()[name = string("op_234_strides_0"), val = tensor([1, 1])]; tensor var_234_pad_0 = const()[name = string("op_234_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_234_dilations_0 = const()[name = string("op_234_dilations_0"), val = tensor([1, 1])]; int32 var_234_groups_0 = const()[name = string("op_234_groups_0"), val = int32(1)]; tensor var_236_weight_0_to_fp16 = const()[name = string("op_236_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(24070208)))]; tensor var_236_bias_0_to_fp16 = const()[name = string("op_236_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(26167424)))]; tensor var_236_cast_fp16 = conv(bias = var_236_bias_0_to_fp16, dilations = var_234_dilations_0, groups = var_234_groups_0, pad = var_234_pad_0, pad_type = var_234_pad_type_0, strides = var_234_strides_0, weight = var_236_weight_0_to_fp16, x = input_9_cast_fp16)[name = string("op_236_cast_fp16")]; string var_243_pad_type_0 = const()[name = string("op_243_pad_type_0"), val = string("valid")]; tensor var_243_strides_0 = const()[name = string("op_243_strides_0"), val = tensor([1, 1])]; tensor var_243_pad_0 = const()[name = string("op_243_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_243_dilations_0 = const()[name = string("op_243_dilations_0"), val = tensor([1, 1])]; int32 var_243_groups_0 = const()[name = string("op_243_groups_0"), val = int32(1)]; tensor layers_0_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_0_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(26169536)))]; tensor layers_0_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_0_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(28266752)))]; tensor var_243_cast_fp16 = conv(bias = layers_0_self_attn_k_proj_bias_to_fp16, dilations = var_243_dilations_0, groups = var_243_groups_0, pad = var_243_pad_0, pad_type = var_243_pad_type_0, strides = var_243_strides_0, weight = layers_0_self_attn_k_proj_weight_to_fp16, x = input_9_cast_fp16)[name = string("op_243_cast_fp16")]; string var_250_pad_type_0 = const()[name = string("op_250_pad_type_0"), val = string("valid")]; tensor var_250_strides_0 = const()[name = string("op_250_strides_0"), val = tensor([1, 1])]; tensor var_250_pad_0 = const()[name = string("op_250_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_250_dilations_0 = const()[name = string("op_250_dilations_0"), val = tensor([1, 1])]; int32 var_250_groups_0 = const()[name = string("op_250_groups_0"), val = int32(1)]; tensor layers_0_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_0_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(28268864)))]; tensor layers_0_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_0_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(30366080)))]; tensor var_250_cast_fp16 = conv(bias = layers_0_self_attn_v_proj_bias_to_fp16, dilations = var_250_dilations_0, groups = var_250_groups_0, pad = var_250_pad_0, pad_type = var_250_pad_type_0, strides = var_250_strides_0, weight = layers_0_self_attn_v_proj_weight_to_fp16, x = input_9_cast_fp16)[name = string("op_250_cast_fp16")]; tensor tile_0 = const()[name = string("tile_0"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(30368192)))]; int32 var_251_axis_0 = const()[name = string("op_251_axis_0"), val = int32(1)]; tensor var_251_cast_fp16_0, tensor var_251_cast_fp16_1, tensor var_251_cast_fp16_2, tensor var_251_cast_fp16_3, tensor var_251_cast_fp16_4, tensor var_251_cast_fp16_5, tensor var_251_cast_fp16_6, tensor var_251_cast_fp16_7, tensor var_251_cast_fp16_8, tensor var_251_cast_fp16_9, tensor var_251_cast_fp16_10, tensor var_251_cast_fp16_11, tensor var_251_cast_fp16_12, tensor var_251_cast_fp16_13, tensor var_251_cast_fp16_14, tensor var_251_cast_fp16_15 = split(axis = var_251_axis_0, split_sizes = tile_0, x = var_236_cast_fp16)[name = string("op_251_cast_fp16")]; tensor tile_1 = const()[name = string("tile_1"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(30368320)))]; int32 var_268_axis_0 = const()[name = string("op_268_axis_0"), val = int32(1)]; tensor var_268_cast_fp16_0, tensor var_268_cast_fp16_1, tensor var_268_cast_fp16_2, tensor var_268_cast_fp16_3, tensor var_268_cast_fp16_4, tensor var_268_cast_fp16_5, tensor var_268_cast_fp16_6, tensor var_268_cast_fp16_7, tensor var_268_cast_fp16_8, tensor var_268_cast_fp16_9, tensor var_268_cast_fp16_10, tensor var_268_cast_fp16_11, tensor var_268_cast_fp16_12, tensor var_268_cast_fp16_13, tensor var_268_cast_fp16_14, tensor var_268_cast_fp16_15 = split(axis = var_268_axis_0, split_sizes = tile_1, x = var_243_cast_fp16)[name = string("op_268_cast_fp16")]; tensor tile_2 = const()[name = string("tile_2"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(30368448)))]; int32 var_285_axis_0 = const()[name = string("op_285_axis_0"), val = int32(1)]; tensor var_285_cast_fp16_0, tensor var_285_cast_fp16_1, tensor var_285_cast_fp16_2, tensor var_285_cast_fp16_3, tensor var_285_cast_fp16_4, tensor var_285_cast_fp16_5, tensor var_285_cast_fp16_6, tensor var_285_cast_fp16_7, tensor var_285_cast_fp16_8, tensor var_285_cast_fp16_9, tensor var_285_cast_fp16_10, tensor var_285_cast_fp16_11, tensor var_285_cast_fp16_12, tensor var_285_cast_fp16_13, tensor var_285_cast_fp16_14, tensor var_285_cast_fp16_15 = split(axis = var_285_axis_0, split_sizes = tile_2, x = var_250_cast_fp16)[name = string("op_285_cast_fp16")]; tensor transpose_0_perm_0 = const()[name = string("transpose_0_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_4 = const()[name = string("concat_4"), val = tensor([1, 104, 64])]; tensor transpose_0_cast_fp16 = transpose(perm = transpose_0_perm_0, x = var_251_cast_fp16_0)[name = string("transpose_4224")]; tensor reshape_0_cast_fp16 = reshape(shape = concat_4, x = transpose_0_cast_fp16)[name = string("reshape_0_cast_fp16")]; tensor transpose_1_perm_0 = const()[name = string("transpose_1_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_5 = const()[name = string("concat_5"), val = tensor([1, 64, 104])]; tensor transpose_1_cast_fp16 = transpose(perm = transpose_1_perm_0, x = var_268_cast_fp16_0)[name = string("transpose_4223")]; tensor reshape_1_cast_fp16 = reshape(shape = concat_5, x = transpose_1_cast_fp16)[name = string("reshape_1_cast_fp16")]; bool matmul_0_transpose_x_0 = const()[name = string("matmul_0_transpose_x_0"), val = bool(false)]; bool matmul_0_transpose_y_0 = const()[name = string("matmul_0_transpose_y_0"), val = bool(false)]; tensor matmul_0_cast_fp16 = matmul(transpose_x = matmul_0_transpose_x_0, transpose_y = matmul_0_transpose_y_0, x = reshape_0_cast_fp16, y = reshape_1_cast_fp16)[name = string("matmul_0_cast_fp16")]; tensor concat_9 = const()[name = string("concat_9"), val = tensor([1, 1, 104, 104])]; tensor reshape_2_cast_fp16 = reshape(shape = concat_9, x = matmul_0_cast_fp16)[name = string("reshape_2_cast_fp16")]; tensor transpose_2304_perm_0 = const()[name = string("transpose_2304_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2305_perm_0 = const()[name = string("transpose_2305_perm_0"), val = tensor([0, 3, 2, 1])]; tensor transpose_2305 = transpose(perm = transpose_2305_perm_0, x = key_bias)[name = string("transpose_4221")]; tensor transpose_2304 = transpose(perm = transpose_2304_perm_0, x = reshape_2_cast_fp16)[name = string("transpose_4222")]; tensor w_3_cast_fp16 = add(x = transpose_2304, y = transpose_2305)[name = string("w_3_cast_fp16")]; tensor var_307_cast_fp16 = softmax(axis = var_195, x = w_3_cast_fp16)[name = string("op_307_cast_fp16")]; string var_309_equation_0 = const()[name = string("op_309_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_309_cast_fp16 = einsum(equation = var_309_equation_0, values = (var_285_cast_fp16_0, var_307_cast_fp16))[name = string("op_309_cast_fp16")]; tensor transpose_2_perm_0 = const()[name = string("transpose_2_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_14 = const()[name = string("concat_14"), val = tensor([1, 104, 64])]; tensor transpose_2_cast_fp16 = transpose(perm = transpose_2_perm_0, x = var_251_cast_fp16_1)[name = string("transpose_4220")]; tensor reshape_3_cast_fp16 = reshape(shape = concat_14, x = transpose_2_cast_fp16)[name = string("reshape_3_cast_fp16")]; tensor transpose_3_perm_0 = const()[name = string("transpose_3_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_15 = const()[name = string("concat_15"), val = tensor([1, 64, 104])]; tensor transpose_3_cast_fp16 = transpose(perm = transpose_3_perm_0, x = var_268_cast_fp16_1)[name = string("transpose_4219")]; tensor reshape_4_cast_fp16 = reshape(shape = concat_15, x = transpose_3_cast_fp16)[name = string("reshape_4_cast_fp16")]; bool matmul_1_transpose_x_0 = const()[name = string("matmul_1_transpose_x_0"), val = bool(false)]; bool matmul_1_transpose_y_0 = const()[name = string("matmul_1_transpose_y_0"), val = bool(false)]; tensor matmul_1_cast_fp16 = matmul(transpose_x = matmul_1_transpose_x_0, transpose_y = matmul_1_transpose_y_0, x = reshape_3_cast_fp16, y = reshape_4_cast_fp16)[name = string("matmul_1_cast_fp16")]; tensor concat_19 = const()[name = string("concat_19"), val = tensor([1, 1, 104, 104])]; tensor reshape_5_cast_fp16 = reshape(shape = concat_19, x = matmul_1_cast_fp16)[name = string("reshape_5_cast_fp16")]; tensor transpose_2689_perm_0 = const()[name = string("transpose_2689_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2689 = transpose(perm = transpose_2689_perm_0, x = reshape_5_cast_fp16)[name = string("transpose_4218")]; tensor w_7_cast_fp16 = add(x = transpose_2689, y = transpose_2305)[name = string("w_7_cast_fp16")]; tensor var_315_cast_fp16 = softmax(axis = var_195, x = w_7_cast_fp16)[name = string("op_315_cast_fp16")]; string var_317_equation_0 = const()[name = string("op_317_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_317_cast_fp16 = einsum(equation = var_317_equation_0, values = (var_285_cast_fp16_1, var_315_cast_fp16))[name = string("op_317_cast_fp16")]; tensor transpose_4_perm_0 = const()[name = string("transpose_4_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_24 = const()[name = string("concat_24"), val = tensor([1, 104, 64])]; tensor transpose_4_cast_fp16 = transpose(perm = transpose_4_perm_0, x = var_251_cast_fp16_2)[name = string("transpose_4217")]; tensor reshape_6_cast_fp16 = reshape(shape = concat_24, x = transpose_4_cast_fp16)[name = string("reshape_6_cast_fp16")]; tensor transpose_5_perm_0 = const()[name = string("transpose_5_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_25 = const()[name = string("concat_25"), val = tensor([1, 64, 104])]; tensor transpose_5_cast_fp16 = transpose(perm = transpose_5_perm_0, x = var_268_cast_fp16_2)[name = string("transpose_4216")]; tensor reshape_7_cast_fp16 = reshape(shape = concat_25, x = transpose_5_cast_fp16)[name = string("reshape_7_cast_fp16")]; bool matmul_2_transpose_x_0 = const()[name = string("matmul_2_transpose_x_0"), val = bool(false)]; bool matmul_2_transpose_y_0 = const()[name = string("matmul_2_transpose_y_0"), val = bool(false)]; tensor matmul_2_cast_fp16 = matmul(transpose_x = matmul_2_transpose_x_0, transpose_y = matmul_2_transpose_y_0, x = reshape_6_cast_fp16, y = reshape_7_cast_fp16)[name = string("matmul_2_cast_fp16")]; tensor concat_29 = const()[name = string("concat_29"), val = tensor([1, 1, 104, 104])]; tensor reshape_8_cast_fp16 = reshape(shape = concat_29, x = matmul_2_cast_fp16)[name = string("reshape_8_cast_fp16")]; tensor transpose_2690_perm_0 = const()[name = string("transpose_2690_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2690 = transpose(perm = transpose_2690_perm_0, x = reshape_8_cast_fp16)[name = string("transpose_4215")]; tensor w_11_cast_fp16 = add(x = transpose_2690, y = transpose_2305)[name = string("w_11_cast_fp16")]; tensor var_323_cast_fp16 = softmax(axis = var_195, x = w_11_cast_fp16)[name = string("op_323_cast_fp16")]; string var_325_equation_0 = const()[name = string("op_325_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_325_cast_fp16 = einsum(equation = var_325_equation_0, values = (var_285_cast_fp16_2, var_323_cast_fp16))[name = string("op_325_cast_fp16")]; tensor transpose_6_perm_0 = const()[name = string("transpose_6_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_34 = const()[name = string("concat_34"), val = tensor([1, 104, 64])]; tensor transpose_6_cast_fp16 = transpose(perm = transpose_6_perm_0, x = var_251_cast_fp16_3)[name = string("transpose_4214")]; tensor reshape_9_cast_fp16 = reshape(shape = concat_34, x = transpose_6_cast_fp16)[name = string("reshape_9_cast_fp16")]; tensor transpose_7_perm_0 = const()[name = string("transpose_7_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_35 = const()[name = string("concat_35"), val = tensor([1, 64, 104])]; tensor transpose_7_cast_fp16 = transpose(perm = transpose_7_perm_0, x = var_268_cast_fp16_3)[name = string("transpose_4213")]; tensor reshape_10_cast_fp16 = reshape(shape = concat_35, x = transpose_7_cast_fp16)[name = string("reshape_10_cast_fp16")]; bool matmul_3_transpose_x_0 = const()[name = string("matmul_3_transpose_x_0"), val = bool(false)]; bool matmul_3_transpose_y_0 = const()[name = string("matmul_3_transpose_y_0"), val = bool(false)]; tensor matmul_3_cast_fp16 = matmul(transpose_x = matmul_3_transpose_x_0, transpose_y = matmul_3_transpose_y_0, x = reshape_9_cast_fp16, y = reshape_10_cast_fp16)[name = string("matmul_3_cast_fp16")]; tensor concat_39 = const()[name = string("concat_39"), val = tensor([1, 1, 104, 104])]; tensor reshape_11_cast_fp16 = reshape(shape = concat_39, x = matmul_3_cast_fp16)[name = string("reshape_11_cast_fp16")]; tensor transpose_2691_perm_0 = const()[name = string("transpose_2691_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2691 = transpose(perm = transpose_2691_perm_0, x = reshape_11_cast_fp16)[name = string("transpose_4212")]; tensor w_15_cast_fp16 = add(x = transpose_2691, y = transpose_2305)[name = string("w_15_cast_fp16")]; tensor var_331_cast_fp16 = softmax(axis = var_195, x = w_15_cast_fp16)[name = string("op_331_cast_fp16")]; string var_333_equation_0 = const()[name = string("op_333_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_333_cast_fp16 = einsum(equation = var_333_equation_0, values = (var_285_cast_fp16_3, var_331_cast_fp16))[name = string("op_333_cast_fp16")]; tensor transpose_8_perm_0 = const()[name = string("transpose_8_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_44 = const()[name = string("concat_44"), val = tensor([1, 104, 64])]; tensor transpose_8_cast_fp16 = transpose(perm = transpose_8_perm_0, x = var_251_cast_fp16_4)[name = string("transpose_4211")]; tensor reshape_12_cast_fp16 = reshape(shape = concat_44, x = transpose_8_cast_fp16)[name = string("reshape_12_cast_fp16")]; tensor transpose_9_perm_0 = const()[name = string("transpose_9_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_45 = const()[name = string("concat_45"), val = tensor([1, 64, 104])]; tensor transpose_9_cast_fp16 = transpose(perm = transpose_9_perm_0, x = var_268_cast_fp16_4)[name = string("transpose_4210")]; tensor reshape_13_cast_fp16 = reshape(shape = concat_45, x = transpose_9_cast_fp16)[name = string("reshape_13_cast_fp16")]; bool matmul_4_transpose_x_0 = const()[name = string("matmul_4_transpose_x_0"), val = bool(false)]; bool matmul_4_transpose_y_0 = const()[name = string("matmul_4_transpose_y_0"), val = bool(false)]; tensor matmul_4_cast_fp16 = matmul(transpose_x = matmul_4_transpose_x_0, transpose_y = matmul_4_transpose_y_0, x = reshape_12_cast_fp16, y = reshape_13_cast_fp16)[name = string("matmul_4_cast_fp16")]; tensor concat_49 = const()[name = string("concat_49"), val = tensor([1, 1, 104, 104])]; tensor reshape_14_cast_fp16 = reshape(shape = concat_49, x = matmul_4_cast_fp16)[name = string("reshape_14_cast_fp16")]; tensor transpose_2692_perm_0 = const()[name = string("transpose_2692_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2692 = transpose(perm = transpose_2692_perm_0, x = reshape_14_cast_fp16)[name = string("transpose_4209")]; tensor w_19_cast_fp16 = add(x = transpose_2692, y = transpose_2305)[name = string("w_19_cast_fp16")]; tensor var_339_cast_fp16 = softmax(axis = var_195, x = w_19_cast_fp16)[name = string("op_339_cast_fp16")]; string var_341_equation_0 = const()[name = string("op_341_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_341_cast_fp16 = einsum(equation = var_341_equation_0, values = (var_285_cast_fp16_4, var_339_cast_fp16))[name = string("op_341_cast_fp16")]; tensor transpose_10_perm_0 = const()[name = string("transpose_10_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_54 = const()[name = string("concat_54"), val = tensor([1, 104, 64])]; tensor transpose_10_cast_fp16 = transpose(perm = transpose_10_perm_0, x = var_251_cast_fp16_5)[name = string("transpose_4208")]; tensor reshape_15_cast_fp16 = reshape(shape = concat_54, x = transpose_10_cast_fp16)[name = string("reshape_15_cast_fp16")]; tensor transpose_11_perm_0 = const()[name = string("transpose_11_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_55 = const()[name = string("concat_55"), val = tensor([1, 64, 104])]; tensor transpose_11_cast_fp16 = transpose(perm = transpose_11_perm_0, x = var_268_cast_fp16_5)[name = string("transpose_4207")]; tensor reshape_16_cast_fp16 = reshape(shape = concat_55, x = transpose_11_cast_fp16)[name = string("reshape_16_cast_fp16")]; bool matmul_5_transpose_x_0 = const()[name = string("matmul_5_transpose_x_0"), val = bool(false)]; bool matmul_5_transpose_y_0 = const()[name = string("matmul_5_transpose_y_0"), val = bool(false)]; tensor matmul_5_cast_fp16 = matmul(transpose_x = matmul_5_transpose_x_0, transpose_y = matmul_5_transpose_y_0, x = reshape_15_cast_fp16, y = reshape_16_cast_fp16)[name = string("matmul_5_cast_fp16")]; tensor concat_59 = const()[name = string("concat_59"), val = tensor([1, 1, 104, 104])]; tensor reshape_17_cast_fp16 = reshape(shape = concat_59, x = matmul_5_cast_fp16)[name = string("reshape_17_cast_fp16")]; tensor transpose_2693_perm_0 = const()[name = string("transpose_2693_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2693 = transpose(perm = transpose_2693_perm_0, x = reshape_17_cast_fp16)[name = string("transpose_4206")]; tensor w_23_cast_fp16 = add(x = transpose_2693, y = transpose_2305)[name = string("w_23_cast_fp16")]; tensor var_347_cast_fp16 = softmax(axis = var_195, x = w_23_cast_fp16)[name = string("op_347_cast_fp16")]; string var_349_equation_0 = const()[name = string("op_349_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_349_cast_fp16 = einsum(equation = var_349_equation_0, values = (var_285_cast_fp16_5, var_347_cast_fp16))[name = string("op_349_cast_fp16")]; tensor transpose_12_perm_0 = const()[name = string("transpose_12_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_64 = const()[name = string("concat_64"), val = tensor([1, 104, 64])]; tensor transpose_12_cast_fp16 = transpose(perm = transpose_12_perm_0, x = var_251_cast_fp16_6)[name = string("transpose_4205")]; tensor reshape_18_cast_fp16 = reshape(shape = concat_64, x = transpose_12_cast_fp16)[name = string("reshape_18_cast_fp16")]; tensor transpose_13_perm_0 = const()[name = string("transpose_13_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_65 = const()[name = string("concat_65"), val = tensor([1, 64, 104])]; tensor transpose_13_cast_fp16 = transpose(perm = transpose_13_perm_0, x = var_268_cast_fp16_6)[name = string("transpose_4204")]; tensor reshape_19_cast_fp16 = reshape(shape = concat_65, x = transpose_13_cast_fp16)[name = string("reshape_19_cast_fp16")]; bool matmul_6_transpose_x_0 = const()[name = string("matmul_6_transpose_x_0"), val = bool(false)]; bool matmul_6_transpose_y_0 = const()[name = string("matmul_6_transpose_y_0"), val = bool(false)]; tensor matmul_6_cast_fp16 = matmul(transpose_x = matmul_6_transpose_x_0, transpose_y = matmul_6_transpose_y_0, x = reshape_18_cast_fp16, y = reshape_19_cast_fp16)[name = string("matmul_6_cast_fp16")]; tensor concat_69 = const()[name = string("concat_69"), val = tensor([1, 1, 104, 104])]; tensor reshape_20_cast_fp16 = reshape(shape = concat_69, x = matmul_6_cast_fp16)[name = string("reshape_20_cast_fp16")]; tensor transpose_2694_perm_0 = const()[name = string("transpose_2694_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2694 = transpose(perm = transpose_2694_perm_0, x = reshape_20_cast_fp16)[name = string("transpose_4203")]; tensor w_27_cast_fp16 = add(x = transpose_2694, y = transpose_2305)[name = string("w_27_cast_fp16")]; tensor var_355_cast_fp16 = softmax(axis = var_195, x = w_27_cast_fp16)[name = string("op_355_cast_fp16")]; string var_357_equation_0 = const()[name = string("op_357_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_357_cast_fp16 = einsum(equation = var_357_equation_0, values = (var_285_cast_fp16_6, var_355_cast_fp16))[name = string("op_357_cast_fp16")]; tensor transpose_14_perm_0 = const()[name = string("transpose_14_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_74 = const()[name = string("concat_74"), val = tensor([1, 104, 64])]; tensor transpose_14_cast_fp16 = transpose(perm = transpose_14_perm_0, x = var_251_cast_fp16_7)[name = string("transpose_4202")]; tensor reshape_21_cast_fp16 = reshape(shape = concat_74, x = transpose_14_cast_fp16)[name = string("reshape_21_cast_fp16")]; tensor transpose_15_perm_0 = const()[name = string("transpose_15_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_75 = const()[name = string("concat_75"), val = tensor([1, 64, 104])]; tensor transpose_15_cast_fp16 = transpose(perm = transpose_15_perm_0, x = var_268_cast_fp16_7)[name = string("transpose_4201")]; tensor reshape_22_cast_fp16 = reshape(shape = concat_75, x = transpose_15_cast_fp16)[name = string("reshape_22_cast_fp16")]; bool matmul_7_transpose_x_0 = const()[name = string("matmul_7_transpose_x_0"), val = bool(false)]; bool matmul_7_transpose_y_0 = const()[name = string("matmul_7_transpose_y_0"), val = bool(false)]; tensor matmul_7_cast_fp16 = matmul(transpose_x = matmul_7_transpose_x_0, transpose_y = matmul_7_transpose_y_0, x = reshape_21_cast_fp16, y = reshape_22_cast_fp16)[name = string("matmul_7_cast_fp16")]; tensor concat_79 = const()[name = string("concat_79"), val = tensor([1, 1, 104, 104])]; tensor reshape_23_cast_fp16 = reshape(shape = concat_79, x = matmul_7_cast_fp16)[name = string("reshape_23_cast_fp16")]; tensor transpose_2695_perm_0 = const()[name = string("transpose_2695_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2695 = transpose(perm = transpose_2695_perm_0, x = reshape_23_cast_fp16)[name = string("transpose_4200")]; tensor w_31_cast_fp16 = add(x = transpose_2695, y = transpose_2305)[name = string("w_31_cast_fp16")]; tensor var_363_cast_fp16 = softmax(axis = var_195, x = w_31_cast_fp16)[name = string("op_363_cast_fp16")]; string var_365_equation_0 = const()[name = string("op_365_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_365_cast_fp16 = einsum(equation = var_365_equation_0, values = (var_285_cast_fp16_7, var_363_cast_fp16))[name = string("op_365_cast_fp16")]; tensor transpose_16_perm_0 = const()[name = string("transpose_16_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_84 = const()[name = string("concat_84"), val = tensor([1, 104, 64])]; tensor transpose_16_cast_fp16 = transpose(perm = transpose_16_perm_0, x = var_251_cast_fp16_8)[name = string("transpose_4199")]; tensor reshape_24_cast_fp16 = reshape(shape = concat_84, x = transpose_16_cast_fp16)[name = string("reshape_24_cast_fp16")]; tensor transpose_17_perm_0 = const()[name = string("transpose_17_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_85 = const()[name = string("concat_85"), val = tensor([1, 64, 104])]; tensor transpose_17_cast_fp16 = transpose(perm = transpose_17_perm_0, x = var_268_cast_fp16_8)[name = string("transpose_4198")]; tensor reshape_25_cast_fp16 = reshape(shape = concat_85, x = transpose_17_cast_fp16)[name = string("reshape_25_cast_fp16")]; bool matmul_8_transpose_x_0 = const()[name = string("matmul_8_transpose_x_0"), val = bool(false)]; bool matmul_8_transpose_y_0 = const()[name = string("matmul_8_transpose_y_0"), val = bool(false)]; tensor matmul_8_cast_fp16 = matmul(transpose_x = matmul_8_transpose_x_0, transpose_y = matmul_8_transpose_y_0, x = reshape_24_cast_fp16, y = reshape_25_cast_fp16)[name = string("matmul_8_cast_fp16")]; tensor concat_89 = const()[name = string("concat_89"), val = tensor([1, 1, 104, 104])]; tensor reshape_26_cast_fp16 = reshape(shape = concat_89, x = matmul_8_cast_fp16)[name = string("reshape_26_cast_fp16")]; tensor transpose_2696_perm_0 = const()[name = string("transpose_2696_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2696 = transpose(perm = transpose_2696_perm_0, x = reshape_26_cast_fp16)[name = string("transpose_4197")]; tensor w_35_cast_fp16 = add(x = transpose_2696, y = transpose_2305)[name = string("w_35_cast_fp16")]; tensor var_371_cast_fp16 = softmax(axis = var_195, x = w_35_cast_fp16)[name = string("op_371_cast_fp16")]; string var_373_equation_0 = const()[name = string("op_373_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_373_cast_fp16 = einsum(equation = var_373_equation_0, values = (var_285_cast_fp16_8, var_371_cast_fp16))[name = string("op_373_cast_fp16")]; tensor transpose_18_perm_0 = const()[name = string("transpose_18_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_94 = const()[name = string("concat_94"), val = tensor([1, 104, 64])]; tensor transpose_18_cast_fp16 = transpose(perm = transpose_18_perm_0, x = var_251_cast_fp16_9)[name = string("transpose_4196")]; tensor reshape_27_cast_fp16 = reshape(shape = concat_94, x = transpose_18_cast_fp16)[name = string("reshape_27_cast_fp16")]; tensor transpose_19_perm_0 = const()[name = string("transpose_19_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_95 = const()[name = string("concat_95"), val = tensor([1, 64, 104])]; tensor transpose_19_cast_fp16 = transpose(perm = transpose_19_perm_0, x = var_268_cast_fp16_9)[name = string("transpose_4195")]; tensor reshape_28_cast_fp16 = reshape(shape = concat_95, x = transpose_19_cast_fp16)[name = string("reshape_28_cast_fp16")]; bool matmul_9_transpose_x_0 = const()[name = string("matmul_9_transpose_x_0"), val = bool(false)]; bool matmul_9_transpose_y_0 = const()[name = string("matmul_9_transpose_y_0"), val = bool(false)]; tensor matmul_9_cast_fp16 = matmul(transpose_x = matmul_9_transpose_x_0, transpose_y = matmul_9_transpose_y_0, x = reshape_27_cast_fp16, y = reshape_28_cast_fp16)[name = string("matmul_9_cast_fp16")]; tensor concat_99 = const()[name = string("concat_99"), val = tensor([1, 1, 104, 104])]; tensor reshape_29_cast_fp16 = reshape(shape = concat_99, x = matmul_9_cast_fp16)[name = string("reshape_29_cast_fp16")]; tensor transpose_2697_perm_0 = const()[name = string("transpose_2697_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2697 = transpose(perm = transpose_2697_perm_0, x = reshape_29_cast_fp16)[name = string("transpose_4194")]; tensor w_39_cast_fp16 = add(x = transpose_2697, y = transpose_2305)[name = string("w_39_cast_fp16")]; tensor var_379_cast_fp16 = softmax(axis = var_195, x = w_39_cast_fp16)[name = string("op_379_cast_fp16")]; string var_381_equation_0 = const()[name = string("op_381_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_381_cast_fp16 = einsum(equation = var_381_equation_0, values = (var_285_cast_fp16_9, var_379_cast_fp16))[name = string("op_381_cast_fp16")]; tensor transpose_20_perm_0 = const()[name = string("transpose_20_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_104 = const()[name = string("concat_104"), val = tensor([1, 104, 64])]; tensor transpose_20_cast_fp16 = transpose(perm = transpose_20_perm_0, x = var_251_cast_fp16_10)[name = string("transpose_4193")]; tensor reshape_30_cast_fp16 = reshape(shape = concat_104, x = transpose_20_cast_fp16)[name = string("reshape_30_cast_fp16")]; tensor transpose_21_perm_0 = const()[name = string("transpose_21_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_105 = const()[name = string("concat_105"), val = tensor([1, 64, 104])]; tensor transpose_21_cast_fp16 = transpose(perm = transpose_21_perm_0, x = var_268_cast_fp16_10)[name = string("transpose_4192")]; tensor reshape_31_cast_fp16 = reshape(shape = concat_105, x = transpose_21_cast_fp16)[name = string("reshape_31_cast_fp16")]; bool matmul_10_transpose_x_0 = const()[name = string("matmul_10_transpose_x_0"), val = bool(false)]; bool matmul_10_transpose_y_0 = const()[name = string("matmul_10_transpose_y_0"), val = bool(false)]; tensor matmul_10_cast_fp16 = matmul(transpose_x = matmul_10_transpose_x_0, transpose_y = matmul_10_transpose_y_0, x = reshape_30_cast_fp16, y = reshape_31_cast_fp16)[name = string("matmul_10_cast_fp16")]; tensor concat_109 = const()[name = string("concat_109"), val = tensor([1, 1, 104, 104])]; tensor reshape_32_cast_fp16 = reshape(shape = concat_109, x = matmul_10_cast_fp16)[name = string("reshape_32_cast_fp16")]; tensor transpose_2698_perm_0 = const()[name = string("transpose_2698_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2698 = transpose(perm = transpose_2698_perm_0, x = reshape_32_cast_fp16)[name = string("transpose_4191")]; tensor w_43_cast_fp16 = add(x = transpose_2698, y = transpose_2305)[name = string("w_43_cast_fp16")]; tensor var_387_cast_fp16 = softmax(axis = var_195, x = w_43_cast_fp16)[name = string("op_387_cast_fp16")]; string var_389_equation_0 = const()[name = string("op_389_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_389_cast_fp16 = einsum(equation = var_389_equation_0, values = (var_285_cast_fp16_10, var_387_cast_fp16))[name = string("op_389_cast_fp16")]; tensor transpose_22_perm_0 = const()[name = string("transpose_22_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_114 = const()[name = string("concat_114"), val = tensor([1, 104, 64])]; tensor transpose_22_cast_fp16 = transpose(perm = transpose_22_perm_0, x = var_251_cast_fp16_11)[name = string("transpose_4190")]; tensor reshape_33_cast_fp16 = reshape(shape = concat_114, x = transpose_22_cast_fp16)[name = string("reshape_33_cast_fp16")]; tensor transpose_23_perm_0 = const()[name = string("transpose_23_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_115 = const()[name = string("concat_115"), val = tensor([1, 64, 104])]; tensor transpose_23_cast_fp16 = transpose(perm = transpose_23_perm_0, x = var_268_cast_fp16_11)[name = string("transpose_4189")]; tensor reshape_34_cast_fp16 = reshape(shape = concat_115, x = transpose_23_cast_fp16)[name = string("reshape_34_cast_fp16")]; bool matmul_11_transpose_x_0 = const()[name = string("matmul_11_transpose_x_0"), val = bool(false)]; bool matmul_11_transpose_y_0 = const()[name = string("matmul_11_transpose_y_0"), val = bool(false)]; tensor matmul_11_cast_fp16 = matmul(transpose_x = matmul_11_transpose_x_0, transpose_y = matmul_11_transpose_y_0, x = reshape_33_cast_fp16, y = reshape_34_cast_fp16)[name = string("matmul_11_cast_fp16")]; tensor concat_119 = const()[name = string("concat_119"), val = tensor([1, 1, 104, 104])]; tensor reshape_35_cast_fp16 = reshape(shape = concat_119, x = matmul_11_cast_fp16)[name = string("reshape_35_cast_fp16")]; tensor transpose_2699_perm_0 = const()[name = string("transpose_2699_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2699 = transpose(perm = transpose_2699_perm_0, x = reshape_35_cast_fp16)[name = string("transpose_4188")]; tensor w_47_cast_fp16 = add(x = transpose_2699, y = transpose_2305)[name = string("w_47_cast_fp16")]; tensor var_395_cast_fp16 = softmax(axis = var_195, x = w_47_cast_fp16)[name = string("op_395_cast_fp16")]; string var_397_equation_0 = const()[name = string("op_397_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_397_cast_fp16 = einsum(equation = var_397_equation_0, values = (var_285_cast_fp16_11, var_395_cast_fp16))[name = string("op_397_cast_fp16")]; tensor transpose_24_perm_0 = const()[name = string("transpose_24_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_124 = const()[name = string("concat_124"), val = tensor([1, 104, 64])]; tensor transpose_24_cast_fp16 = transpose(perm = transpose_24_perm_0, x = var_251_cast_fp16_12)[name = string("transpose_4187")]; tensor reshape_36_cast_fp16 = reshape(shape = concat_124, x = transpose_24_cast_fp16)[name = string("reshape_36_cast_fp16")]; tensor transpose_25_perm_0 = const()[name = string("transpose_25_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_125 = const()[name = string("concat_125"), val = tensor([1, 64, 104])]; tensor transpose_25_cast_fp16 = transpose(perm = transpose_25_perm_0, x = var_268_cast_fp16_12)[name = string("transpose_4186")]; tensor reshape_37_cast_fp16 = reshape(shape = concat_125, x = transpose_25_cast_fp16)[name = string("reshape_37_cast_fp16")]; bool matmul_12_transpose_x_0 = const()[name = string("matmul_12_transpose_x_0"), val = bool(false)]; bool matmul_12_transpose_y_0 = const()[name = string("matmul_12_transpose_y_0"), val = bool(false)]; tensor matmul_12_cast_fp16 = matmul(transpose_x = matmul_12_transpose_x_0, transpose_y = matmul_12_transpose_y_0, x = reshape_36_cast_fp16, y = reshape_37_cast_fp16)[name = string("matmul_12_cast_fp16")]; tensor concat_129 = const()[name = string("concat_129"), val = tensor([1, 1, 104, 104])]; tensor reshape_38_cast_fp16 = reshape(shape = concat_129, x = matmul_12_cast_fp16)[name = string("reshape_38_cast_fp16")]; tensor transpose_2700_perm_0 = const()[name = string("transpose_2700_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2700 = transpose(perm = transpose_2700_perm_0, x = reshape_38_cast_fp16)[name = string("transpose_4185")]; tensor w_51_cast_fp16 = add(x = transpose_2700, y = transpose_2305)[name = string("w_51_cast_fp16")]; tensor var_403_cast_fp16 = softmax(axis = var_195, x = w_51_cast_fp16)[name = string("op_403_cast_fp16")]; string var_405_equation_0 = const()[name = string("op_405_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_405_cast_fp16 = einsum(equation = var_405_equation_0, values = (var_285_cast_fp16_12, var_403_cast_fp16))[name = string("op_405_cast_fp16")]; tensor transpose_26_perm_0 = const()[name = string("transpose_26_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_134 = const()[name = string("concat_134"), val = tensor([1, 104, 64])]; tensor transpose_26_cast_fp16 = transpose(perm = transpose_26_perm_0, x = var_251_cast_fp16_13)[name = string("transpose_4184")]; tensor reshape_39_cast_fp16 = reshape(shape = concat_134, x = transpose_26_cast_fp16)[name = string("reshape_39_cast_fp16")]; tensor transpose_27_perm_0 = const()[name = string("transpose_27_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_135 = const()[name = string("concat_135"), val = tensor([1, 64, 104])]; tensor transpose_27_cast_fp16 = transpose(perm = transpose_27_perm_0, x = var_268_cast_fp16_13)[name = string("transpose_4183")]; tensor reshape_40_cast_fp16 = reshape(shape = concat_135, x = transpose_27_cast_fp16)[name = string("reshape_40_cast_fp16")]; bool matmul_13_transpose_x_0 = const()[name = string("matmul_13_transpose_x_0"), val = bool(false)]; bool matmul_13_transpose_y_0 = const()[name = string("matmul_13_transpose_y_0"), val = bool(false)]; tensor matmul_13_cast_fp16 = matmul(transpose_x = matmul_13_transpose_x_0, transpose_y = matmul_13_transpose_y_0, x = reshape_39_cast_fp16, y = reshape_40_cast_fp16)[name = string("matmul_13_cast_fp16")]; tensor concat_139 = const()[name = string("concat_139"), val = tensor([1, 1, 104, 104])]; tensor reshape_41_cast_fp16 = reshape(shape = concat_139, x = matmul_13_cast_fp16)[name = string("reshape_41_cast_fp16")]; tensor transpose_2701_perm_0 = const()[name = string("transpose_2701_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2701 = transpose(perm = transpose_2701_perm_0, x = reshape_41_cast_fp16)[name = string("transpose_4182")]; tensor w_55_cast_fp16 = add(x = transpose_2701, y = transpose_2305)[name = string("w_55_cast_fp16")]; tensor var_411_cast_fp16 = softmax(axis = var_195, x = w_55_cast_fp16)[name = string("op_411_cast_fp16")]; string var_413_equation_0 = const()[name = string("op_413_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_413_cast_fp16 = einsum(equation = var_413_equation_0, values = (var_285_cast_fp16_13, var_411_cast_fp16))[name = string("op_413_cast_fp16")]; tensor transpose_28_perm_0 = const()[name = string("transpose_28_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_144 = const()[name = string("concat_144"), val = tensor([1, 104, 64])]; tensor transpose_28_cast_fp16 = transpose(perm = transpose_28_perm_0, x = var_251_cast_fp16_14)[name = string("transpose_4181")]; tensor reshape_42_cast_fp16 = reshape(shape = concat_144, x = transpose_28_cast_fp16)[name = string("reshape_42_cast_fp16")]; tensor transpose_29_perm_0 = const()[name = string("transpose_29_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_145 = const()[name = string("concat_145"), val = tensor([1, 64, 104])]; tensor transpose_29_cast_fp16 = transpose(perm = transpose_29_perm_0, x = var_268_cast_fp16_14)[name = string("transpose_4180")]; tensor reshape_43_cast_fp16 = reshape(shape = concat_145, x = transpose_29_cast_fp16)[name = string("reshape_43_cast_fp16")]; bool matmul_14_transpose_x_0 = const()[name = string("matmul_14_transpose_x_0"), val = bool(false)]; bool matmul_14_transpose_y_0 = const()[name = string("matmul_14_transpose_y_0"), val = bool(false)]; tensor matmul_14_cast_fp16 = matmul(transpose_x = matmul_14_transpose_x_0, transpose_y = matmul_14_transpose_y_0, x = reshape_42_cast_fp16, y = reshape_43_cast_fp16)[name = string("matmul_14_cast_fp16")]; tensor concat_149 = const()[name = string("concat_149"), val = tensor([1, 1, 104, 104])]; tensor reshape_44_cast_fp16 = reshape(shape = concat_149, x = matmul_14_cast_fp16)[name = string("reshape_44_cast_fp16")]; tensor transpose_2702_perm_0 = const()[name = string("transpose_2702_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2702 = transpose(perm = transpose_2702_perm_0, x = reshape_44_cast_fp16)[name = string("transpose_4179")]; tensor w_59_cast_fp16 = add(x = transpose_2702, y = transpose_2305)[name = string("w_59_cast_fp16")]; tensor var_419_cast_fp16 = softmax(axis = var_195, x = w_59_cast_fp16)[name = string("op_419_cast_fp16")]; string var_421_equation_0 = const()[name = string("op_421_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_421_cast_fp16 = einsum(equation = var_421_equation_0, values = (var_285_cast_fp16_14, var_419_cast_fp16))[name = string("op_421_cast_fp16")]; tensor transpose_30_perm_0 = const()[name = string("transpose_30_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_154 = const()[name = string("concat_154"), val = tensor([1, 104, 64])]; tensor transpose_30_cast_fp16 = transpose(perm = transpose_30_perm_0, x = var_251_cast_fp16_15)[name = string("transpose_4178")]; tensor reshape_45_cast_fp16 = reshape(shape = concat_154, x = transpose_30_cast_fp16)[name = string("reshape_45_cast_fp16")]; tensor transpose_31_perm_0 = const()[name = string("transpose_31_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_155 = const()[name = string("concat_155"), val = tensor([1, 64, 104])]; tensor transpose_31_cast_fp16 = transpose(perm = transpose_31_perm_0, x = var_268_cast_fp16_15)[name = string("transpose_4177")]; tensor reshape_46_cast_fp16 = reshape(shape = concat_155, x = transpose_31_cast_fp16)[name = string("reshape_46_cast_fp16")]; bool matmul_15_transpose_x_0 = const()[name = string("matmul_15_transpose_x_0"), val = bool(false)]; bool matmul_15_transpose_y_0 = const()[name = string("matmul_15_transpose_y_0"), val = bool(false)]; tensor matmul_15_cast_fp16 = matmul(transpose_x = matmul_15_transpose_x_0, transpose_y = matmul_15_transpose_y_0, x = reshape_45_cast_fp16, y = reshape_46_cast_fp16)[name = string("matmul_15_cast_fp16")]; tensor concat_159 = const()[name = string("concat_159"), val = tensor([1, 1, 104, 104])]; tensor reshape_47_cast_fp16 = reshape(shape = concat_159, x = matmul_15_cast_fp16)[name = string("reshape_47_cast_fp16")]; tensor transpose_2703_perm_0 = const()[name = string("transpose_2703_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2703 = transpose(perm = transpose_2703_perm_0, x = reshape_47_cast_fp16)[name = string("transpose_4176")]; tensor w_63_cast_fp16 = add(x = transpose_2703, y = transpose_2305)[name = string("w_63_cast_fp16")]; tensor var_427_cast_fp16 = softmax(axis = var_195, x = w_63_cast_fp16)[name = string("op_427_cast_fp16")]; string var_429_equation_0 = const()[name = string("op_429_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_429_cast_fp16 = einsum(equation = var_429_equation_0, values = (var_285_cast_fp16_15, var_427_cast_fp16))[name = string("op_429_cast_fp16")]; bool input_11_interleave_0 = const()[name = string("input_11_interleave_0"), val = bool(false)]; tensor input_11_cast_fp16 = concat(axis = var_195, interleave = input_11_interleave_0, values = (var_309_cast_fp16, var_317_cast_fp16, var_325_cast_fp16, var_333_cast_fp16, var_341_cast_fp16, var_349_cast_fp16, var_357_cast_fp16, var_365_cast_fp16, var_373_cast_fp16, var_381_cast_fp16, var_389_cast_fp16, var_397_cast_fp16, var_405_cast_fp16, var_413_cast_fp16, var_421_cast_fp16, var_429_cast_fp16))[name = string("input_11_cast_fp16")]; string var_438_pad_type_0 = const()[name = string("op_438_pad_type_0"), val = string("valid")]; tensor var_438_strides_0 = const()[name = string("op_438_strides_0"), val = tensor([1, 1])]; tensor var_438_pad_0 = const()[name = string("op_438_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_438_dilations_0 = const()[name = string("op_438_dilations_0"), val = tensor([1, 1])]; int32 var_438_groups_0 = const()[name = string("op_438_groups_0"), val = int32(1)]; tensor layers_0_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_0_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(30368576)))]; tensor layers_0_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_0_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(32465792)))]; tensor var_438_cast_fp16 = conv(bias = layers_0_self_attn_out_proj_bias_to_fp16, dilations = var_438_dilations_0, groups = var_438_groups_0, pad = var_438_pad_0, pad_type = var_438_pad_type_0, strides = var_438_strides_0, weight = layers_0_self_attn_out_proj_weight_to_fp16, x = input_11_cast_fp16)[name = string("op_438_cast_fp16")]; tensor x_17_cast_fp16 = add(x = x_13_cast_fp16, y = var_438_cast_fp16)[name = string("x_17_cast_fp16")]; tensor mu_3_axes_0 = const()[name = string("mu_3_axes_0"), val = tensor([1])]; bool mu_3_keep_dims_0 = const()[name = string("mu_3_keep_dims_0"), val = bool(true)]; tensor mu_3_cast_fp16 = reduce_mean(axes = mu_3_axes_0, keep_dims = mu_3_keep_dims_0, x = x_17_cast_fp16)[name = string("mu_3_cast_fp16")]; tensor var_444_cast_fp16 = sub(x = x_17_cast_fp16, y = mu_3_cast_fp16)[name = string("op_444_cast_fp16")]; fp16 var_198_promoted_1_to_fp16 = const()[name = string("op_198_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_445_cast_fp16 = pow(x = var_444_cast_fp16, y = var_198_promoted_1_to_fp16)[name = string("op_445_cast_fp16")]; tensor var_3_axes_0 = const()[name = string("var_3_axes_0"), val = tensor([1])]; bool var_3_keep_dims_0 = const()[name = string("var_3_keep_dims_0"), val = bool(true)]; tensor var_3_cast_fp16 = reduce_mean(axes = var_3_axes_0, keep_dims = var_3_keep_dims_0, x = var_445_cast_fp16)[name = string("var_3_cast_fp16")]; fp16 var_449_to_fp16 = const()[name = string("op_449_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_450_cast_fp16 = add(x = var_3_cast_fp16, y = var_449_to_fp16)[name = string("op_450_cast_fp16")]; fp32 var_451_epsilon_0 = const()[name = string("op_451_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_451_cast_fp16 = rsqrt(epsilon = var_451_epsilon_0, x = var_450_cast_fp16)[name = string("op_451_cast_fp16")]; tensor x_19_cast_fp16 = mul(x = var_444_cast_fp16, y = var_451_cast_fp16)[name = string("x_19_cast_fp16")]; tensor input_13_gamma_0_to_fp16 = const()[name = string("input_13_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(32467904)))]; tensor input_13_beta_0_to_fp16 = const()[name = string("input_13_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(32470016)))]; fp16 input_13_epsilon_0_to_fp16 = const()[name = string("input_13_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_13_cast_fp16 = batch_norm(beta = input_13_beta_0_to_fp16, epsilon = input_13_epsilon_0_to_fp16, gamma = input_13_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_19_cast_fp16)[name = string("input_13_cast_fp16")]; string x_21_pad_type_0 = const()[name = string("x_21_pad_type_0"), val = string("valid")]; tensor x_21_strides_0 = const()[name = string("x_21_strides_0"), val = tensor([1, 1])]; tensor x_21_pad_0 = const()[name = string("x_21_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_21_dilations_0 = const()[name = string("x_21_dilations_0"), val = tensor([1, 1])]; int32 x_21_groups_0 = const()[name = string("x_21_groups_0"), val = int32(1)]; tensor layers_0_fc1_weight_to_fp16 = const()[name = string("layers_0_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(32472128)))]; tensor layers_0_fc1_bias_to_fp16 = const()[name = string("layers_0_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(40860800)))]; tensor x_21_cast_fp16 = conv(bias = layers_0_fc1_bias_to_fp16, dilations = x_21_dilations_0, groups = x_21_groups_0, pad = x_21_pad_0, pad_type = x_21_pad_type_0, strides = x_21_strides_0, weight = layers_0_fc1_weight_to_fp16, x = input_13_cast_fp16)[name = string("x_21_cast_fp16")]; fp16 var_466_to_fp16 = const()[name = string("op_466_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_467_cast_fp16 = mul(x = x_21_cast_fp16, y = var_466_to_fp16)[name = string("op_467_cast_fp16")]; tensor var_468_cast_fp16 = mul(x = var_467_cast_fp16, y = x_21_cast_fp16)[name = string("op_468_cast_fp16")]; tensor var_469_cast_fp16 = mul(x = var_468_cast_fp16, y = x_21_cast_fp16)[name = string("op_469_cast_fp16")]; tensor var_470_cast_fp16 = add(x = x_21_cast_fp16, y = var_469_cast_fp16)[name = string("op_470_cast_fp16")]; fp16 var_471_to_fp16 = const()[name = string("op_471_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_7_cast_fp16 = mul(x = var_470_cast_fp16, y = var_471_to_fp16)[name = string("u_7_cast_fp16")]; fp16 var_473_to_fp16 = const()[name = string("op_473_to_fp16"), val = fp16(0x1p-1)]; tensor var_474_cast_fp16 = mul(x = x_21_cast_fp16, y = var_473_to_fp16)[name = string("op_474_cast_fp16")]; tensor var_475_cast_fp16 = tanh(x = u_7_cast_fp16)[name = string("op_475_cast_fp16")]; fp16 var_476_to_fp16 = const()[name = string("op_476_to_fp16"), val = fp16(0x1p+0)]; tensor var_477_cast_fp16 = add(x = var_475_cast_fp16, y = var_476_to_fp16)[name = string("op_477_cast_fp16")]; tensor input_15_cast_fp16 = mul(x = var_474_cast_fp16, y = var_477_cast_fp16)[name = string("input_15_cast_fp16")]; string h_1_pad_type_0 = const()[name = string("h_1_pad_type_0"), val = string("valid")]; tensor h_1_strides_0 = const()[name = string("h_1_strides_0"), val = tensor([1, 1])]; tensor h_1_pad_0 = const()[name = string("h_1_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_1_dilations_0 = const()[name = string("h_1_dilations_0"), val = tensor([1, 1])]; int32 h_1_groups_0 = const()[name = string("h_1_groups_0"), val = int32(1)]; tensor layers_0_fc2_weight_to_fp16 = const()[name = string("layers_0_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(40869056)))]; tensor layers_0_fc2_bias_to_fp16 = const()[name = string("layers_0_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(49257728)))]; tensor h_1_cast_fp16 = conv(bias = layers_0_fc2_bias_to_fp16, dilations = h_1_dilations_0, groups = h_1_groups_0, pad = h_1_pad_0, pad_type = h_1_pad_type_0, strides = h_1_strides_0, weight = layers_0_fc2_weight_to_fp16, x = input_15_cast_fp16)[name = string("h_1_cast_fp16")]; tensor x_23_cast_fp16 = add(x = x_17_cast_fp16, y = h_1_cast_fp16)[name = string("x_23_cast_fp16")]; int32 var_493 = const()[name = string("op_493"), val = int32(1)]; tensor mu_5_axes_0 = const()[name = string("mu_5_axes_0"), val = tensor([1])]; bool mu_5_keep_dims_0 = const()[name = string("mu_5_keep_dims_0"), val = bool(true)]; tensor mu_5_cast_fp16 = reduce_mean(axes = mu_5_axes_0, keep_dims = mu_5_keep_dims_0, x = x_23_cast_fp16)[name = string("mu_5_cast_fp16")]; tensor var_507_cast_fp16 = sub(x = x_23_cast_fp16, y = mu_5_cast_fp16)[name = string("op_507_cast_fp16")]; fp16 var_496_promoted_to_fp16 = const()[name = string("op_496_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_508_cast_fp16 = pow(x = var_507_cast_fp16, y = var_496_promoted_to_fp16)[name = string("op_508_cast_fp16")]; tensor var_5_axes_0 = const()[name = string("var_5_axes_0"), val = tensor([1])]; bool var_5_keep_dims_0 = const()[name = string("var_5_keep_dims_0"), val = bool(true)]; tensor var_5_cast_fp16 = reduce_mean(axes = var_5_axes_0, keep_dims = var_5_keep_dims_0, x = var_508_cast_fp16)[name = string("var_5_cast_fp16")]; fp16 var_512_to_fp16 = const()[name = string("op_512_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_513_cast_fp16 = add(x = var_5_cast_fp16, y = var_512_to_fp16)[name = string("op_513_cast_fp16")]; fp32 var_514_epsilon_0 = const()[name = string("op_514_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_514_cast_fp16 = rsqrt(epsilon = var_514_epsilon_0, x = var_513_cast_fp16)[name = string("op_514_cast_fp16")]; tensor x_25_cast_fp16 = mul(x = var_507_cast_fp16, y = var_514_cast_fp16)[name = string("x_25_cast_fp16")]; tensor input_17_gamma_0_to_fp16 = const()[name = string("input_17_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(49259840)))]; tensor input_17_beta_0_to_fp16 = const()[name = string("input_17_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(49261952)))]; fp16 input_17_epsilon_0_to_fp16 = const()[name = string("input_17_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_17_cast_fp16 = batch_norm(beta = input_17_beta_0_to_fp16, epsilon = input_17_epsilon_0_to_fp16, gamma = input_17_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_25_cast_fp16)[name = string("input_17_cast_fp16")]; string var_532_pad_type_0 = const()[name = string("op_532_pad_type_0"), val = string("valid")]; tensor var_532_strides_0 = const()[name = string("op_532_strides_0"), val = tensor([1, 1])]; tensor var_532_pad_0 = const()[name = string("op_532_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_532_dilations_0 = const()[name = string("op_532_dilations_0"), val = tensor([1, 1])]; int32 var_532_groups_0 = const()[name = string("op_532_groups_0"), val = int32(1)]; tensor var_534_weight_0_to_fp16 = const()[name = string("op_534_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(49264064)))]; tensor var_534_bias_0_to_fp16 = const()[name = string("op_534_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(51361280)))]; tensor var_534_cast_fp16 = conv(bias = var_534_bias_0_to_fp16, dilations = var_532_dilations_0, groups = var_532_groups_0, pad = var_532_pad_0, pad_type = var_532_pad_type_0, strides = var_532_strides_0, weight = var_534_weight_0_to_fp16, x = input_17_cast_fp16)[name = string("op_534_cast_fp16")]; string var_541_pad_type_0 = const()[name = string("op_541_pad_type_0"), val = string("valid")]; tensor var_541_strides_0 = const()[name = string("op_541_strides_0"), val = tensor([1, 1])]; tensor var_541_pad_0 = const()[name = string("op_541_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_541_dilations_0 = const()[name = string("op_541_dilations_0"), val = tensor([1, 1])]; int32 var_541_groups_0 = const()[name = string("op_541_groups_0"), val = int32(1)]; tensor layers_1_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_1_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(51363392)))]; tensor layers_1_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_1_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(53460608)))]; tensor var_541_cast_fp16 = conv(bias = layers_1_self_attn_k_proj_bias_to_fp16, dilations = var_541_dilations_0, groups = var_541_groups_0, pad = var_541_pad_0, pad_type = var_541_pad_type_0, strides = var_541_strides_0, weight = layers_1_self_attn_k_proj_weight_to_fp16, x = input_17_cast_fp16)[name = string("op_541_cast_fp16")]; string var_548_pad_type_0 = const()[name = string("op_548_pad_type_0"), val = string("valid")]; tensor var_548_strides_0 = const()[name = string("op_548_strides_0"), val = tensor([1, 1])]; tensor var_548_pad_0 = const()[name = string("op_548_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_548_dilations_0 = const()[name = string("op_548_dilations_0"), val = tensor([1, 1])]; int32 var_548_groups_0 = const()[name = string("op_548_groups_0"), val = int32(1)]; tensor layers_1_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_1_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(53462720)))]; tensor layers_1_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_1_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(55559936)))]; tensor var_548_cast_fp16 = conv(bias = layers_1_self_attn_v_proj_bias_to_fp16, dilations = var_548_dilations_0, groups = var_548_groups_0, pad = var_548_pad_0, pad_type = var_548_pad_type_0, strides = var_548_strides_0, weight = layers_1_self_attn_v_proj_weight_to_fp16, x = input_17_cast_fp16)[name = string("op_548_cast_fp16")]; tensor tile_3 = const()[name = string("tile_3"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(55562048)))]; int32 var_549_axis_0 = const()[name = string("op_549_axis_0"), val = int32(1)]; tensor var_549_cast_fp16_0, tensor var_549_cast_fp16_1, tensor var_549_cast_fp16_2, tensor var_549_cast_fp16_3, tensor var_549_cast_fp16_4, tensor var_549_cast_fp16_5, tensor var_549_cast_fp16_6, tensor var_549_cast_fp16_7, tensor var_549_cast_fp16_8, tensor var_549_cast_fp16_9, tensor var_549_cast_fp16_10, tensor var_549_cast_fp16_11, tensor var_549_cast_fp16_12, tensor var_549_cast_fp16_13, tensor var_549_cast_fp16_14, tensor var_549_cast_fp16_15 = split(axis = var_549_axis_0, split_sizes = tile_3, x = var_534_cast_fp16)[name = string("op_549_cast_fp16")]; tensor tile_4 = const()[name = string("tile_4"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(55562176)))]; int32 var_566_axis_0 = const()[name = string("op_566_axis_0"), val = int32(1)]; tensor var_566_cast_fp16_0, tensor var_566_cast_fp16_1, tensor var_566_cast_fp16_2, tensor var_566_cast_fp16_3, tensor var_566_cast_fp16_4, tensor var_566_cast_fp16_5, tensor var_566_cast_fp16_6, tensor var_566_cast_fp16_7, tensor var_566_cast_fp16_8, tensor var_566_cast_fp16_9, tensor var_566_cast_fp16_10, tensor var_566_cast_fp16_11, tensor var_566_cast_fp16_12, tensor var_566_cast_fp16_13, tensor var_566_cast_fp16_14, tensor var_566_cast_fp16_15 = split(axis = var_566_axis_0, split_sizes = tile_4, x = var_541_cast_fp16)[name = string("op_566_cast_fp16")]; tensor tile_5 = const()[name = string("tile_5"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(55562304)))]; int32 var_583_axis_0 = const()[name = string("op_583_axis_0"), val = int32(1)]; tensor var_583_cast_fp16_0, tensor var_583_cast_fp16_1, tensor var_583_cast_fp16_2, tensor var_583_cast_fp16_3, tensor var_583_cast_fp16_4, tensor var_583_cast_fp16_5, tensor var_583_cast_fp16_6, tensor var_583_cast_fp16_7, tensor var_583_cast_fp16_8, tensor var_583_cast_fp16_9, tensor var_583_cast_fp16_10, tensor var_583_cast_fp16_11, tensor var_583_cast_fp16_12, tensor var_583_cast_fp16_13, tensor var_583_cast_fp16_14, tensor var_583_cast_fp16_15 = split(axis = var_583_axis_0, split_sizes = tile_5, x = var_548_cast_fp16)[name = string("op_583_cast_fp16")]; tensor transpose_32_perm_0 = const()[name = string("transpose_32_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_164 = const()[name = string("concat_164"), val = tensor([1, 104, 64])]; tensor transpose_32_cast_fp16 = transpose(perm = transpose_32_perm_0, x = var_549_cast_fp16_0)[name = string("transpose_4175")]; tensor reshape_48_cast_fp16 = reshape(shape = concat_164, x = transpose_32_cast_fp16)[name = string("reshape_48_cast_fp16")]; tensor transpose_33_perm_0 = const()[name = string("transpose_33_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_165 = const()[name = string("concat_165"), val = tensor([1, 64, 104])]; tensor transpose_33_cast_fp16 = transpose(perm = transpose_33_perm_0, x = var_566_cast_fp16_0)[name = string("transpose_4174")]; tensor reshape_49_cast_fp16 = reshape(shape = concat_165, x = transpose_33_cast_fp16)[name = string("reshape_49_cast_fp16")]; bool matmul_16_transpose_x_0 = const()[name = string("matmul_16_transpose_x_0"), val = bool(false)]; bool matmul_16_transpose_y_0 = const()[name = string("matmul_16_transpose_y_0"), val = bool(false)]; tensor matmul_16_cast_fp16 = matmul(transpose_x = matmul_16_transpose_x_0, transpose_y = matmul_16_transpose_y_0, x = reshape_48_cast_fp16, y = reshape_49_cast_fp16)[name = string("matmul_16_cast_fp16")]; tensor concat_169 = const()[name = string("concat_169"), val = tensor([1, 1, 104, 104])]; tensor reshape_50_cast_fp16 = reshape(shape = concat_169, x = matmul_16_cast_fp16)[name = string("reshape_50_cast_fp16")]; tensor transpose_2704_perm_0 = const()[name = string("transpose_2704_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2704 = transpose(perm = transpose_2704_perm_0, x = reshape_50_cast_fp16)[name = string("transpose_4173")]; tensor w_67_cast_fp16 = add(x = transpose_2704, y = transpose_2305)[name = string("w_67_cast_fp16")]; tensor var_605_cast_fp16 = softmax(axis = var_493, x = w_67_cast_fp16)[name = string("op_605_cast_fp16")]; string var_607_equation_0 = const()[name = string("op_607_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_607_cast_fp16 = einsum(equation = var_607_equation_0, values = (var_583_cast_fp16_0, var_605_cast_fp16))[name = string("op_607_cast_fp16")]; tensor transpose_34_perm_0 = const()[name = string("transpose_34_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_174 = const()[name = string("concat_174"), val = tensor([1, 104, 64])]; tensor transpose_34_cast_fp16 = transpose(perm = transpose_34_perm_0, x = var_549_cast_fp16_1)[name = string("transpose_4172")]; tensor reshape_51_cast_fp16 = reshape(shape = concat_174, x = transpose_34_cast_fp16)[name = string("reshape_51_cast_fp16")]; tensor transpose_35_perm_0 = const()[name = string("transpose_35_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_175 = const()[name = string("concat_175"), val = tensor([1, 64, 104])]; tensor transpose_35_cast_fp16 = transpose(perm = transpose_35_perm_0, x = var_566_cast_fp16_1)[name = string("transpose_4171")]; tensor reshape_52_cast_fp16 = reshape(shape = concat_175, x = transpose_35_cast_fp16)[name = string("reshape_52_cast_fp16")]; bool matmul_17_transpose_x_0 = const()[name = string("matmul_17_transpose_x_0"), val = bool(false)]; bool matmul_17_transpose_y_0 = const()[name = string("matmul_17_transpose_y_0"), val = bool(false)]; tensor matmul_17_cast_fp16 = matmul(transpose_x = matmul_17_transpose_x_0, transpose_y = matmul_17_transpose_y_0, x = reshape_51_cast_fp16, y = reshape_52_cast_fp16)[name = string("matmul_17_cast_fp16")]; tensor concat_179 = const()[name = string("concat_179"), val = tensor([1, 1, 104, 104])]; tensor reshape_53_cast_fp16 = reshape(shape = concat_179, x = matmul_17_cast_fp16)[name = string("reshape_53_cast_fp16")]; tensor transpose_2705_perm_0 = const()[name = string("transpose_2705_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2705 = transpose(perm = transpose_2705_perm_0, x = reshape_53_cast_fp16)[name = string("transpose_4170")]; tensor w_71_cast_fp16 = add(x = transpose_2705, y = transpose_2305)[name = string("w_71_cast_fp16")]; tensor var_613_cast_fp16 = softmax(axis = var_493, x = w_71_cast_fp16)[name = string("op_613_cast_fp16")]; string var_615_equation_0 = const()[name = string("op_615_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_615_cast_fp16 = einsum(equation = var_615_equation_0, values = (var_583_cast_fp16_1, var_613_cast_fp16))[name = string("op_615_cast_fp16")]; tensor transpose_36_perm_0 = const()[name = string("transpose_36_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_184 = const()[name = string("concat_184"), val = tensor([1, 104, 64])]; tensor transpose_36_cast_fp16 = transpose(perm = transpose_36_perm_0, x = var_549_cast_fp16_2)[name = string("transpose_4169")]; tensor reshape_54_cast_fp16 = reshape(shape = concat_184, x = transpose_36_cast_fp16)[name = string("reshape_54_cast_fp16")]; tensor transpose_37_perm_0 = const()[name = string("transpose_37_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_185 = const()[name = string("concat_185"), val = tensor([1, 64, 104])]; tensor transpose_37_cast_fp16 = transpose(perm = transpose_37_perm_0, x = var_566_cast_fp16_2)[name = string("transpose_4168")]; tensor reshape_55_cast_fp16 = reshape(shape = concat_185, x = transpose_37_cast_fp16)[name = string("reshape_55_cast_fp16")]; bool matmul_18_transpose_x_0 = const()[name = string("matmul_18_transpose_x_0"), val = bool(false)]; bool matmul_18_transpose_y_0 = const()[name = string("matmul_18_transpose_y_0"), val = bool(false)]; tensor matmul_18_cast_fp16 = matmul(transpose_x = matmul_18_transpose_x_0, transpose_y = matmul_18_transpose_y_0, x = reshape_54_cast_fp16, y = reshape_55_cast_fp16)[name = string("matmul_18_cast_fp16")]; tensor concat_189 = const()[name = string("concat_189"), val = tensor([1, 1, 104, 104])]; tensor reshape_56_cast_fp16 = reshape(shape = concat_189, x = matmul_18_cast_fp16)[name = string("reshape_56_cast_fp16")]; tensor transpose_2706_perm_0 = const()[name = string("transpose_2706_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2706 = transpose(perm = transpose_2706_perm_0, x = reshape_56_cast_fp16)[name = string("transpose_4167")]; tensor w_75_cast_fp16 = add(x = transpose_2706, y = transpose_2305)[name = string("w_75_cast_fp16")]; tensor var_621_cast_fp16 = softmax(axis = var_493, x = w_75_cast_fp16)[name = string("op_621_cast_fp16")]; string var_623_equation_0 = const()[name = string("op_623_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_623_cast_fp16 = einsum(equation = var_623_equation_0, values = (var_583_cast_fp16_2, var_621_cast_fp16))[name = string("op_623_cast_fp16")]; tensor transpose_38_perm_0 = const()[name = string("transpose_38_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_194 = const()[name = string("concat_194"), val = tensor([1, 104, 64])]; tensor transpose_38_cast_fp16 = transpose(perm = transpose_38_perm_0, x = var_549_cast_fp16_3)[name = string("transpose_4166")]; tensor reshape_57_cast_fp16 = reshape(shape = concat_194, x = transpose_38_cast_fp16)[name = string("reshape_57_cast_fp16")]; tensor transpose_39_perm_0 = const()[name = string("transpose_39_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_195 = const()[name = string("concat_195"), val = tensor([1, 64, 104])]; tensor transpose_39_cast_fp16 = transpose(perm = transpose_39_perm_0, x = var_566_cast_fp16_3)[name = string("transpose_4165")]; tensor reshape_58_cast_fp16 = reshape(shape = concat_195, x = transpose_39_cast_fp16)[name = string("reshape_58_cast_fp16")]; bool matmul_19_transpose_x_0 = const()[name = string("matmul_19_transpose_x_0"), val = bool(false)]; bool matmul_19_transpose_y_0 = const()[name = string("matmul_19_transpose_y_0"), val = bool(false)]; tensor matmul_19_cast_fp16 = matmul(transpose_x = matmul_19_transpose_x_0, transpose_y = matmul_19_transpose_y_0, x = reshape_57_cast_fp16, y = reshape_58_cast_fp16)[name = string("matmul_19_cast_fp16")]; tensor concat_199 = const()[name = string("concat_199"), val = tensor([1, 1, 104, 104])]; tensor reshape_59_cast_fp16 = reshape(shape = concat_199, x = matmul_19_cast_fp16)[name = string("reshape_59_cast_fp16")]; tensor transpose_2707_perm_0 = const()[name = string("transpose_2707_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2707 = transpose(perm = transpose_2707_perm_0, x = reshape_59_cast_fp16)[name = string("transpose_4164")]; tensor w_79_cast_fp16 = add(x = transpose_2707, y = transpose_2305)[name = string("w_79_cast_fp16")]; tensor var_629_cast_fp16 = softmax(axis = var_493, x = w_79_cast_fp16)[name = string("op_629_cast_fp16")]; string var_631_equation_0 = const()[name = string("op_631_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_631_cast_fp16 = einsum(equation = var_631_equation_0, values = (var_583_cast_fp16_3, var_629_cast_fp16))[name = string("op_631_cast_fp16")]; tensor transpose_40_perm_0 = const()[name = string("transpose_40_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_204 = const()[name = string("concat_204"), val = tensor([1, 104, 64])]; tensor transpose_40_cast_fp16 = transpose(perm = transpose_40_perm_0, x = var_549_cast_fp16_4)[name = string("transpose_4163")]; tensor reshape_60_cast_fp16 = reshape(shape = concat_204, x = transpose_40_cast_fp16)[name = string("reshape_60_cast_fp16")]; tensor transpose_41_perm_0 = const()[name = string("transpose_41_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_205 = const()[name = string("concat_205"), val = tensor([1, 64, 104])]; tensor transpose_41_cast_fp16 = transpose(perm = transpose_41_perm_0, x = var_566_cast_fp16_4)[name = string("transpose_4162")]; tensor reshape_61_cast_fp16 = reshape(shape = concat_205, x = transpose_41_cast_fp16)[name = string("reshape_61_cast_fp16")]; bool matmul_20_transpose_x_0 = const()[name = string("matmul_20_transpose_x_0"), val = bool(false)]; bool matmul_20_transpose_y_0 = const()[name = string("matmul_20_transpose_y_0"), val = bool(false)]; tensor matmul_20_cast_fp16 = matmul(transpose_x = matmul_20_transpose_x_0, transpose_y = matmul_20_transpose_y_0, x = reshape_60_cast_fp16, y = reshape_61_cast_fp16)[name = string("matmul_20_cast_fp16")]; tensor concat_209 = const()[name = string("concat_209"), val = tensor([1, 1, 104, 104])]; tensor reshape_62_cast_fp16 = reshape(shape = concat_209, x = matmul_20_cast_fp16)[name = string("reshape_62_cast_fp16")]; tensor transpose_2708_perm_0 = const()[name = string("transpose_2708_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2708 = transpose(perm = transpose_2708_perm_0, x = reshape_62_cast_fp16)[name = string("transpose_4161")]; tensor w_83_cast_fp16 = add(x = transpose_2708, y = transpose_2305)[name = string("w_83_cast_fp16")]; tensor var_637_cast_fp16 = softmax(axis = var_493, x = w_83_cast_fp16)[name = string("op_637_cast_fp16")]; string var_639_equation_0 = const()[name = string("op_639_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_639_cast_fp16 = einsum(equation = var_639_equation_0, values = (var_583_cast_fp16_4, var_637_cast_fp16))[name = string("op_639_cast_fp16")]; tensor transpose_42_perm_0 = const()[name = string("transpose_42_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_214 = const()[name = string("concat_214"), val = tensor([1, 104, 64])]; tensor transpose_42_cast_fp16 = transpose(perm = transpose_42_perm_0, x = var_549_cast_fp16_5)[name = string("transpose_4160")]; tensor reshape_63_cast_fp16 = reshape(shape = concat_214, x = transpose_42_cast_fp16)[name = string("reshape_63_cast_fp16")]; tensor transpose_43_perm_0 = const()[name = string("transpose_43_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_215 = const()[name = string("concat_215"), val = tensor([1, 64, 104])]; tensor transpose_43_cast_fp16 = transpose(perm = transpose_43_perm_0, x = var_566_cast_fp16_5)[name = string("transpose_4159")]; tensor reshape_64_cast_fp16 = reshape(shape = concat_215, x = transpose_43_cast_fp16)[name = string("reshape_64_cast_fp16")]; bool matmul_21_transpose_x_0 = const()[name = string("matmul_21_transpose_x_0"), val = bool(false)]; bool matmul_21_transpose_y_0 = const()[name = string("matmul_21_transpose_y_0"), val = bool(false)]; tensor matmul_21_cast_fp16 = matmul(transpose_x = matmul_21_transpose_x_0, transpose_y = matmul_21_transpose_y_0, x = reshape_63_cast_fp16, y = reshape_64_cast_fp16)[name = string("matmul_21_cast_fp16")]; tensor concat_219 = const()[name = string("concat_219"), val = tensor([1, 1, 104, 104])]; tensor reshape_65_cast_fp16 = reshape(shape = concat_219, x = matmul_21_cast_fp16)[name = string("reshape_65_cast_fp16")]; tensor transpose_2709_perm_0 = const()[name = string("transpose_2709_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2709 = transpose(perm = transpose_2709_perm_0, x = reshape_65_cast_fp16)[name = string("transpose_4158")]; tensor w_87_cast_fp16 = add(x = transpose_2709, y = transpose_2305)[name = string("w_87_cast_fp16")]; tensor var_645_cast_fp16 = softmax(axis = var_493, x = w_87_cast_fp16)[name = string("op_645_cast_fp16")]; string var_647_equation_0 = const()[name = string("op_647_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_647_cast_fp16 = einsum(equation = var_647_equation_0, values = (var_583_cast_fp16_5, var_645_cast_fp16))[name = string("op_647_cast_fp16")]; tensor transpose_44_perm_0 = const()[name = string("transpose_44_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_224 = const()[name = string("concat_224"), val = tensor([1, 104, 64])]; tensor transpose_44_cast_fp16 = transpose(perm = transpose_44_perm_0, x = var_549_cast_fp16_6)[name = string("transpose_4157")]; tensor reshape_66_cast_fp16 = reshape(shape = concat_224, x = transpose_44_cast_fp16)[name = string("reshape_66_cast_fp16")]; tensor transpose_45_perm_0 = const()[name = string("transpose_45_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_225 = const()[name = string("concat_225"), val = tensor([1, 64, 104])]; tensor transpose_45_cast_fp16 = transpose(perm = transpose_45_perm_0, x = var_566_cast_fp16_6)[name = string("transpose_4156")]; tensor reshape_67_cast_fp16 = reshape(shape = concat_225, x = transpose_45_cast_fp16)[name = string("reshape_67_cast_fp16")]; bool matmul_22_transpose_x_0 = const()[name = string("matmul_22_transpose_x_0"), val = bool(false)]; bool matmul_22_transpose_y_0 = const()[name = string("matmul_22_transpose_y_0"), val = bool(false)]; tensor matmul_22_cast_fp16 = matmul(transpose_x = matmul_22_transpose_x_0, transpose_y = matmul_22_transpose_y_0, x = reshape_66_cast_fp16, y = reshape_67_cast_fp16)[name = string("matmul_22_cast_fp16")]; tensor concat_229 = const()[name = string("concat_229"), val = tensor([1, 1, 104, 104])]; tensor reshape_68_cast_fp16 = reshape(shape = concat_229, x = matmul_22_cast_fp16)[name = string("reshape_68_cast_fp16")]; tensor transpose_2710_perm_0 = const()[name = string("transpose_2710_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2710 = transpose(perm = transpose_2710_perm_0, x = reshape_68_cast_fp16)[name = string("transpose_4155")]; tensor w_91_cast_fp16 = add(x = transpose_2710, y = transpose_2305)[name = string("w_91_cast_fp16")]; tensor var_653_cast_fp16 = softmax(axis = var_493, x = w_91_cast_fp16)[name = string("op_653_cast_fp16")]; string var_655_equation_0 = const()[name = string("op_655_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_655_cast_fp16 = einsum(equation = var_655_equation_0, values = (var_583_cast_fp16_6, var_653_cast_fp16))[name = string("op_655_cast_fp16")]; tensor transpose_46_perm_0 = const()[name = string("transpose_46_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_234 = const()[name = string("concat_234"), val = tensor([1, 104, 64])]; tensor transpose_46_cast_fp16 = transpose(perm = transpose_46_perm_0, x = var_549_cast_fp16_7)[name = string("transpose_4154")]; tensor reshape_69_cast_fp16 = reshape(shape = concat_234, x = transpose_46_cast_fp16)[name = string("reshape_69_cast_fp16")]; tensor transpose_47_perm_0 = const()[name = string("transpose_47_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_235 = const()[name = string("concat_235"), val = tensor([1, 64, 104])]; tensor transpose_47_cast_fp16 = transpose(perm = transpose_47_perm_0, x = var_566_cast_fp16_7)[name = string("transpose_4153")]; tensor reshape_70_cast_fp16 = reshape(shape = concat_235, x = transpose_47_cast_fp16)[name = string("reshape_70_cast_fp16")]; bool matmul_23_transpose_x_0 = const()[name = string("matmul_23_transpose_x_0"), val = bool(false)]; bool matmul_23_transpose_y_0 = const()[name = string("matmul_23_transpose_y_0"), val = bool(false)]; tensor matmul_23_cast_fp16 = matmul(transpose_x = matmul_23_transpose_x_0, transpose_y = matmul_23_transpose_y_0, x = reshape_69_cast_fp16, y = reshape_70_cast_fp16)[name = string("matmul_23_cast_fp16")]; tensor concat_239 = const()[name = string("concat_239"), val = tensor([1, 1, 104, 104])]; tensor reshape_71_cast_fp16 = reshape(shape = concat_239, x = matmul_23_cast_fp16)[name = string("reshape_71_cast_fp16")]; tensor transpose_2711_perm_0 = const()[name = string("transpose_2711_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2711 = transpose(perm = transpose_2711_perm_0, x = reshape_71_cast_fp16)[name = string("transpose_4152")]; tensor w_95_cast_fp16 = add(x = transpose_2711, y = transpose_2305)[name = string("w_95_cast_fp16")]; tensor var_661_cast_fp16 = softmax(axis = var_493, x = w_95_cast_fp16)[name = string("op_661_cast_fp16")]; string var_663_equation_0 = const()[name = string("op_663_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_663_cast_fp16 = einsum(equation = var_663_equation_0, values = (var_583_cast_fp16_7, var_661_cast_fp16))[name = string("op_663_cast_fp16")]; tensor transpose_48_perm_0 = const()[name = string("transpose_48_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_244 = const()[name = string("concat_244"), val = tensor([1, 104, 64])]; tensor transpose_48_cast_fp16 = transpose(perm = transpose_48_perm_0, x = var_549_cast_fp16_8)[name = string("transpose_4151")]; tensor reshape_72_cast_fp16 = reshape(shape = concat_244, x = transpose_48_cast_fp16)[name = string("reshape_72_cast_fp16")]; tensor transpose_49_perm_0 = const()[name = string("transpose_49_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_245 = const()[name = string("concat_245"), val = tensor([1, 64, 104])]; tensor transpose_49_cast_fp16 = transpose(perm = transpose_49_perm_0, x = var_566_cast_fp16_8)[name = string("transpose_4150")]; tensor reshape_73_cast_fp16 = reshape(shape = concat_245, x = transpose_49_cast_fp16)[name = string("reshape_73_cast_fp16")]; bool matmul_24_transpose_x_0 = const()[name = string("matmul_24_transpose_x_0"), val = bool(false)]; bool matmul_24_transpose_y_0 = const()[name = string("matmul_24_transpose_y_0"), val = bool(false)]; tensor matmul_24_cast_fp16 = matmul(transpose_x = matmul_24_transpose_x_0, transpose_y = matmul_24_transpose_y_0, x = reshape_72_cast_fp16, y = reshape_73_cast_fp16)[name = string("matmul_24_cast_fp16")]; tensor concat_249 = const()[name = string("concat_249"), val = tensor([1, 1, 104, 104])]; tensor reshape_74_cast_fp16 = reshape(shape = concat_249, x = matmul_24_cast_fp16)[name = string("reshape_74_cast_fp16")]; tensor transpose_2712_perm_0 = const()[name = string("transpose_2712_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2712 = transpose(perm = transpose_2712_perm_0, x = reshape_74_cast_fp16)[name = string("transpose_4149")]; tensor w_99_cast_fp16 = add(x = transpose_2712, y = transpose_2305)[name = string("w_99_cast_fp16")]; tensor var_669_cast_fp16 = softmax(axis = var_493, x = w_99_cast_fp16)[name = string("op_669_cast_fp16")]; string var_671_equation_0 = const()[name = string("op_671_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_671_cast_fp16 = einsum(equation = var_671_equation_0, values = (var_583_cast_fp16_8, var_669_cast_fp16))[name = string("op_671_cast_fp16")]; tensor transpose_50_perm_0 = const()[name = string("transpose_50_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_254 = const()[name = string("concat_254"), val = tensor([1, 104, 64])]; tensor transpose_50_cast_fp16 = transpose(perm = transpose_50_perm_0, x = var_549_cast_fp16_9)[name = string("transpose_4148")]; tensor reshape_75_cast_fp16 = reshape(shape = concat_254, x = transpose_50_cast_fp16)[name = string("reshape_75_cast_fp16")]; tensor transpose_51_perm_0 = const()[name = string("transpose_51_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_255 = const()[name = string("concat_255"), val = tensor([1, 64, 104])]; tensor transpose_51_cast_fp16 = transpose(perm = transpose_51_perm_0, x = var_566_cast_fp16_9)[name = string("transpose_4147")]; tensor reshape_76_cast_fp16 = reshape(shape = concat_255, x = transpose_51_cast_fp16)[name = string("reshape_76_cast_fp16")]; bool matmul_25_transpose_x_0 = const()[name = string("matmul_25_transpose_x_0"), val = bool(false)]; bool matmul_25_transpose_y_0 = const()[name = string("matmul_25_transpose_y_0"), val = bool(false)]; tensor matmul_25_cast_fp16 = matmul(transpose_x = matmul_25_transpose_x_0, transpose_y = matmul_25_transpose_y_0, x = reshape_75_cast_fp16, y = reshape_76_cast_fp16)[name = string("matmul_25_cast_fp16")]; tensor concat_259 = const()[name = string("concat_259"), val = tensor([1, 1, 104, 104])]; tensor reshape_77_cast_fp16 = reshape(shape = concat_259, x = matmul_25_cast_fp16)[name = string("reshape_77_cast_fp16")]; tensor transpose_2713_perm_0 = const()[name = string("transpose_2713_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2713 = transpose(perm = transpose_2713_perm_0, x = reshape_77_cast_fp16)[name = string("transpose_4146")]; tensor w_103_cast_fp16 = add(x = transpose_2713, y = transpose_2305)[name = string("w_103_cast_fp16")]; tensor var_677_cast_fp16 = softmax(axis = var_493, x = w_103_cast_fp16)[name = string("op_677_cast_fp16")]; string var_679_equation_0 = const()[name = string("op_679_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_679_cast_fp16 = einsum(equation = var_679_equation_0, values = (var_583_cast_fp16_9, var_677_cast_fp16))[name = string("op_679_cast_fp16")]; tensor transpose_52_perm_0 = const()[name = string("transpose_52_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_264 = const()[name = string("concat_264"), val = tensor([1, 104, 64])]; tensor transpose_52_cast_fp16 = transpose(perm = transpose_52_perm_0, x = var_549_cast_fp16_10)[name = string("transpose_4145")]; tensor reshape_78_cast_fp16 = reshape(shape = concat_264, x = transpose_52_cast_fp16)[name = string("reshape_78_cast_fp16")]; tensor transpose_53_perm_0 = const()[name = string("transpose_53_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_265 = const()[name = string("concat_265"), val = tensor([1, 64, 104])]; tensor transpose_53_cast_fp16 = transpose(perm = transpose_53_perm_0, x = var_566_cast_fp16_10)[name = string("transpose_4144")]; tensor reshape_79_cast_fp16 = reshape(shape = concat_265, x = transpose_53_cast_fp16)[name = string("reshape_79_cast_fp16")]; bool matmul_26_transpose_x_0 = const()[name = string("matmul_26_transpose_x_0"), val = bool(false)]; bool matmul_26_transpose_y_0 = const()[name = string("matmul_26_transpose_y_0"), val = bool(false)]; tensor matmul_26_cast_fp16 = matmul(transpose_x = matmul_26_transpose_x_0, transpose_y = matmul_26_transpose_y_0, x = reshape_78_cast_fp16, y = reshape_79_cast_fp16)[name = string("matmul_26_cast_fp16")]; tensor concat_269 = const()[name = string("concat_269"), val = tensor([1, 1, 104, 104])]; tensor reshape_80_cast_fp16 = reshape(shape = concat_269, x = matmul_26_cast_fp16)[name = string("reshape_80_cast_fp16")]; tensor transpose_2714_perm_0 = const()[name = string("transpose_2714_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2714 = transpose(perm = transpose_2714_perm_0, x = reshape_80_cast_fp16)[name = string("transpose_4143")]; tensor w_107_cast_fp16 = add(x = transpose_2714, y = transpose_2305)[name = string("w_107_cast_fp16")]; tensor var_685_cast_fp16 = softmax(axis = var_493, x = w_107_cast_fp16)[name = string("op_685_cast_fp16")]; string var_687_equation_0 = const()[name = string("op_687_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_687_cast_fp16 = einsum(equation = var_687_equation_0, values = (var_583_cast_fp16_10, var_685_cast_fp16))[name = string("op_687_cast_fp16")]; tensor transpose_54_perm_0 = const()[name = string("transpose_54_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_274 = const()[name = string("concat_274"), val = tensor([1, 104, 64])]; tensor transpose_54_cast_fp16 = transpose(perm = transpose_54_perm_0, x = var_549_cast_fp16_11)[name = string("transpose_4142")]; tensor reshape_81_cast_fp16 = reshape(shape = concat_274, x = transpose_54_cast_fp16)[name = string("reshape_81_cast_fp16")]; tensor transpose_55_perm_0 = const()[name = string("transpose_55_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_275 = const()[name = string("concat_275"), val = tensor([1, 64, 104])]; tensor transpose_55_cast_fp16 = transpose(perm = transpose_55_perm_0, x = var_566_cast_fp16_11)[name = string("transpose_4141")]; tensor reshape_82_cast_fp16 = reshape(shape = concat_275, x = transpose_55_cast_fp16)[name = string("reshape_82_cast_fp16")]; bool matmul_27_transpose_x_0 = const()[name = string("matmul_27_transpose_x_0"), val = bool(false)]; bool matmul_27_transpose_y_0 = const()[name = string("matmul_27_transpose_y_0"), val = bool(false)]; tensor matmul_27_cast_fp16 = matmul(transpose_x = matmul_27_transpose_x_0, transpose_y = matmul_27_transpose_y_0, x = reshape_81_cast_fp16, y = reshape_82_cast_fp16)[name = string("matmul_27_cast_fp16")]; tensor concat_279 = const()[name = string("concat_279"), val = tensor([1, 1, 104, 104])]; tensor reshape_83_cast_fp16 = reshape(shape = concat_279, x = matmul_27_cast_fp16)[name = string("reshape_83_cast_fp16")]; tensor transpose_2715_perm_0 = const()[name = string("transpose_2715_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2715 = transpose(perm = transpose_2715_perm_0, x = reshape_83_cast_fp16)[name = string("transpose_4140")]; tensor w_111_cast_fp16 = add(x = transpose_2715, y = transpose_2305)[name = string("w_111_cast_fp16")]; tensor var_693_cast_fp16 = softmax(axis = var_493, x = w_111_cast_fp16)[name = string("op_693_cast_fp16")]; string var_695_equation_0 = const()[name = string("op_695_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_695_cast_fp16 = einsum(equation = var_695_equation_0, values = (var_583_cast_fp16_11, var_693_cast_fp16))[name = string("op_695_cast_fp16")]; tensor transpose_56_perm_0 = const()[name = string("transpose_56_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_284 = const()[name = string("concat_284"), val = tensor([1, 104, 64])]; tensor transpose_56_cast_fp16 = transpose(perm = transpose_56_perm_0, x = var_549_cast_fp16_12)[name = string("transpose_4139")]; tensor reshape_84_cast_fp16 = reshape(shape = concat_284, x = transpose_56_cast_fp16)[name = string("reshape_84_cast_fp16")]; tensor transpose_57_perm_0 = const()[name = string("transpose_57_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_285 = const()[name = string("concat_285"), val = tensor([1, 64, 104])]; tensor transpose_57_cast_fp16 = transpose(perm = transpose_57_perm_0, x = var_566_cast_fp16_12)[name = string("transpose_4138")]; tensor reshape_85_cast_fp16 = reshape(shape = concat_285, x = transpose_57_cast_fp16)[name = string("reshape_85_cast_fp16")]; bool matmul_28_transpose_x_0 = const()[name = string("matmul_28_transpose_x_0"), val = bool(false)]; bool matmul_28_transpose_y_0 = const()[name = string("matmul_28_transpose_y_0"), val = bool(false)]; tensor matmul_28_cast_fp16 = matmul(transpose_x = matmul_28_transpose_x_0, transpose_y = matmul_28_transpose_y_0, x = reshape_84_cast_fp16, y = reshape_85_cast_fp16)[name = string("matmul_28_cast_fp16")]; tensor concat_289 = const()[name = string("concat_289"), val = tensor([1, 1, 104, 104])]; tensor reshape_86_cast_fp16 = reshape(shape = concat_289, x = matmul_28_cast_fp16)[name = string("reshape_86_cast_fp16")]; tensor transpose_2716_perm_0 = const()[name = string("transpose_2716_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2716 = transpose(perm = transpose_2716_perm_0, x = reshape_86_cast_fp16)[name = string("transpose_4137")]; tensor w_115_cast_fp16 = add(x = transpose_2716, y = transpose_2305)[name = string("w_115_cast_fp16")]; tensor var_701_cast_fp16 = softmax(axis = var_493, x = w_115_cast_fp16)[name = string("op_701_cast_fp16")]; string var_703_equation_0 = const()[name = string("op_703_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_703_cast_fp16 = einsum(equation = var_703_equation_0, values = (var_583_cast_fp16_12, var_701_cast_fp16))[name = string("op_703_cast_fp16")]; tensor transpose_58_perm_0 = const()[name = string("transpose_58_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_294 = const()[name = string("concat_294"), val = tensor([1, 104, 64])]; tensor transpose_58_cast_fp16 = transpose(perm = transpose_58_perm_0, x = var_549_cast_fp16_13)[name = string("transpose_4136")]; tensor reshape_87_cast_fp16 = reshape(shape = concat_294, x = transpose_58_cast_fp16)[name = string("reshape_87_cast_fp16")]; tensor transpose_59_perm_0 = const()[name = string("transpose_59_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_295 = const()[name = string("concat_295"), val = tensor([1, 64, 104])]; tensor transpose_59_cast_fp16 = transpose(perm = transpose_59_perm_0, x = var_566_cast_fp16_13)[name = string("transpose_4135")]; tensor reshape_88_cast_fp16 = reshape(shape = concat_295, x = transpose_59_cast_fp16)[name = string("reshape_88_cast_fp16")]; bool matmul_29_transpose_x_0 = const()[name = string("matmul_29_transpose_x_0"), val = bool(false)]; bool matmul_29_transpose_y_0 = const()[name = string("matmul_29_transpose_y_0"), val = bool(false)]; tensor matmul_29_cast_fp16 = matmul(transpose_x = matmul_29_transpose_x_0, transpose_y = matmul_29_transpose_y_0, x = reshape_87_cast_fp16, y = reshape_88_cast_fp16)[name = string("matmul_29_cast_fp16")]; tensor concat_299 = const()[name = string("concat_299"), val = tensor([1, 1, 104, 104])]; tensor reshape_89_cast_fp16 = reshape(shape = concat_299, x = matmul_29_cast_fp16)[name = string("reshape_89_cast_fp16")]; tensor transpose_2717_perm_0 = const()[name = string("transpose_2717_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2717 = transpose(perm = transpose_2717_perm_0, x = reshape_89_cast_fp16)[name = string("transpose_4134")]; tensor w_119_cast_fp16 = add(x = transpose_2717, y = transpose_2305)[name = string("w_119_cast_fp16")]; tensor var_709_cast_fp16 = softmax(axis = var_493, x = w_119_cast_fp16)[name = string("op_709_cast_fp16")]; string var_711_equation_0 = const()[name = string("op_711_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_711_cast_fp16 = einsum(equation = var_711_equation_0, values = (var_583_cast_fp16_13, var_709_cast_fp16))[name = string("op_711_cast_fp16")]; tensor transpose_60_perm_0 = const()[name = string("transpose_60_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_304 = const()[name = string("concat_304"), val = tensor([1, 104, 64])]; tensor transpose_60_cast_fp16 = transpose(perm = transpose_60_perm_0, x = var_549_cast_fp16_14)[name = string("transpose_4133")]; tensor reshape_90_cast_fp16 = reshape(shape = concat_304, x = transpose_60_cast_fp16)[name = string("reshape_90_cast_fp16")]; tensor transpose_61_perm_0 = const()[name = string("transpose_61_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_305 = const()[name = string("concat_305"), val = tensor([1, 64, 104])]; tensor transpose_61_cast_fp16 = transpose(perm = transpose_61_perm_0, x = var_566_cast_fp16_14)[name = string("transpose_4132")]; tensor reshape_91_cast_fp16 = reshape(shape = concat_305, x = transpose_61_cast_fp16)[name = string("reshape_91_cast_fp16")]; bool matmul_30_transpose_x_0 = const()[name = string("matmul_30_transpose_x_0"), val = bool(false)]; bool matmul_30_transpose_y_0 = const()[name = string("matmul_30_transpose_y_0"), val = bool(false)]; tensor matmul_30_cast_fp16 = matmul(transpose_x = matmul_30_transpose_x_0, transpose_y = matmul_30_transpose_y_0, x = reshape_90_cast_fp16, y = reshape_91_cast_fp16)[name = string("matmul_30_cast_fp16")]; tensor concat_309 = const()[name = string("concat_309"), val = tensor([1, 1, 104, 104])]; tensor reshape_92_cast_fp16 = reshape(shape = concat_309, x = matmul_30_cast_fp16)[name = string("reshape_92_cast_fp16")]; tensor transpose_2718_perm_0 = const()[name = string("transpose_2718_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2718 = transpose(perm = transpose_2718_perm_0, x = reshape_92_cast_fp16)[name = string("transpose_4131")]; tensor w_123_cast_fp16 = add(x = transpose_2718, y = transpose_2305)[name = string("w_123_cast_fp16")]; tensor var_717_cast_fp16 = softmax(axis = var_493, x = w_123_cast_fp16)[name = string("op_717_cast_fp16")]; string var_719_equation_0 = const()[name = string("op_719_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_719_cast_fp16 = einsum(equation = var_719_equation_0, values = (var_583_cast_fp16_14, var_717_cast_fp16))[name = string("op_719_cast_fp16")]; tensor transpose_62_perm_0 = const()[name = string("transpose_62_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_314 = const()[name = string("concat_314"), val = tensor([1, 104, 64])]; tensor transpose_62_cast_fp16 = transpose(perm = transpose_62_perm_0, x = var_549_cast_fp16_15)[name = string("transpose_4130")]; tensor reshape_93_cast_fp16 = reshape(shape = concat_314, x = transpose_62_cast_fp16)[name = string("reshape_93_cast_fp16")]; tensor transpose_63_perm_0 = const()[name = string("transpose_63_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_315 = const()[name = string("concat_315"), val = tensor([1, 64, 104])]; tensor transpose_63_cast_fp16 = transpose(perm = transpose_63_perm_0, x = var_566_cast_fp16_15)[name = string("transpose_4129")]; tensor reshape_94_cast_fp16 = reshape(shape = concat_315, x = transpose_63_cast_fp16)[name = string("reshape_94_cast_fp16")]; bool matmul_31_transpose_x_0 = const()[name = string("matmul_31_transpose_x_0"), val = bool(false)]; bool matmul_31_transpose_y_0 = const()[name = string("matmul_31_transpose_y_0"), val = bool(false)]; tensor matmul_31_cast_fp16 = matmul(transpose_x = matmul_31_transpose_x_0, transpose_y = matmul_31_transpose_y_0, x = reshape_93_cast_fp16, y = reshape_94_cast_fp16)[name = string("matmul_31_cast_fp16")]; tensor concat_319 = const()[name = string("concat_319"), val = tensor([1, 1, 104, 104])]; tensor reshape_95_cast_fp16 = reshape(shape = concat_319, x = matmul_31_cast_fp16)[name = string("reshape_95_cast_fp16")]; tensor transpose_2719_perm_0 = const()[name = string("transpose_2719_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2719 = transpose(perm = transpose_2719_perm_0, x = reshape_95_cast_fp16)[name = string("transpose_4128")]; tensor w_127_cast_fp16 = add(x = transpose_2719, y = transpose_2305)[name = string("w_127_cast_fp16")]; tensor var_725_cast_fp16 = softmax(axis = var_493, x = w_127_cast_fp16)[name = string("op_725_cast_fp16")]; string var_727_equation_0 = const()[name = string("op_727_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_727_cast_fp16 = einsum(equation = var_727_equation_0, values = (var_583_cast_fp16_15, var_725_cast_fp16))[name = string("op_727_cast_fp16")]; bool input_19_interleave_0 = const()[name = string("input_19_interleave_0"), val = bool(false)]; tensor input_19_cast_fp16 = concat(axis = var_493, interleave = input_19_interleave_0, values = (var_607_cast_fp16, var_615_cast_fp16, var_623_cast_fp16, var_631_cast_fp16, var_639_cast_fp16, var_647_cast_fp16, var_655_cast_fp16, var_663_cast_fp16, var_671_cast_fp16, var_679_cast_fp16, var_687_cast_fp16, var_695_cast_fp16, var_703_cast_fp16, var_711_cast_fp16, var_719_cast_fp16, var_727_cast_fp16))[name = string("input_19_cast_fp16")]; string var_736_pad_type_0 = const()[name = string("op_736_pad_type_0"), val = string("valid")]; tensor var_736_strides_0 = const()[name = string("op_736_strides_0"), val = tensor([1, 1])]; tensor var_736_pad_0 = const()[name = string("op_736_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_736_dilations_0 = const()[name = string("op_736_dilations_0"), val = tensor([1, 1])]; int32 var_736_groups_0 = const()[name = string("op_736_groups_0"), val = int32(1)]; tensor layers_1_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_1_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(55562432)))]; tensor layers_1_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_1_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(57659648)))]; tensor var_736_cast_fp16 = conv(bias = layers_1_self_attn_out_proj_bias_to_fp16, dilations = var_736_dilations_0, groups = var_736_groups_0, pad = var_736_pad_0, pad_type = var_736_pad_type_0, strides = var_736_strides_0, weight = layers_1_self_attn_out_proj_weight_to_fp16, x = input_19_cast_fp16)[name = string("op_736_cast_fp16")]; tensor x_27_cast_fp16 = add(x = x_23_cast_fp16, y = var_736_cast_fp16)[name = string("x_27_cast_fp16")]; tensor mu_7_axes_0 = const()[name = string("mu_7_axes_0"), val = tensor([1])]; bool mu_7_keep_dims_0 = const()[name = string("mu_7_keep_dims_0"), val = bool(true)]; tensor mu_7_cast_fp16 = reduce_mean(axes = mu_7_axes_0, keep_dims = mu_7_keep_dims_0, x = x_27_cast_fp16)[name = string("mu_7_cast_fp16")]; tensor var_742_cast_fp16 = sub(x = x_27_cast_fp16, y = mu_7_cast_fp16)[name = string("op_742_cast_fp16")]; fp16 var_496_promoted_1_to_fp16 = const()[name = string("op_496_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_743_cast_fp16 = pow(x = var_742_cast_fp16, y = var_496_promoted_1_to_fp16)[name = string("op_743_cast_fp16")]; tensor var_7_axes_0 = const()[name = string("var_7_axes_0"), val = tensor([1])]; bool var_7_keep_dims_0 = const()[name = string("var_7_keep_dims_0"), val = bool(true)]; tensor var_7_cast_fp16 = reduce_mean(axes = var_7_axes_0, keep_dims = var_7_keep_dims_0, x = var_743_cast_fp16)[name = string("var_7_cast_fp16")]; fp16 var_747_to_fp16 = const()[name = string("op_747_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_748_cast_fp16 = add(x = var_7_cast_fp16, y = var_747_to_fp16)[name = string("op_748_cast_fp16")]; fp32 var_749_epsilon_0 = const()[name = string("op_749_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_749_cast_fp16 = rsqrt(epsilon = var_749_epsilon_0, x = var_748_cast_fp16)[name = string("op_749_cast_fp16")]; tensor x_29_cast_fp16 = mul(x = var_742_cast_fp16, y = var_749_cast_fp16)[name = string("x_29_cast_fp16")]; tensor input_21_gamma_0_to_fp16 = const()[name = string("input_21_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(57661760)))]; tensor input_21_beta_0_to_fp16 = const()[name = string("input_21_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(57663872)))]; fp16 input_21_epsilon_0_to_fp16 = const()[name = string("input_21_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_21_cast_fp16 = batch_norm(beta = input_21_beta_0_to_fp16, epsilon = input_21_epsilon_0_to_fp16, gamma = input_21_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_29_cast_fp16)[name = string("input_21_cast_fp16")]; string x_31_pad_type_0 = const()[name = string("x_31_pad_type_0"), val = string("valid")]; tensor x_31_strides_0 = const()[name = string("x_31_strides_0"), val = tensor([1, 1])]; tensor x_31_pad_0 = const()[name = string("x_31_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_31_dilations_0 = const()[name = string("x_31_dilations_0"), val = tensor([1, 1])]; int32 x_31_groups_0 = const()[name = string("x_31_groups_0"), val = int32(1)]; tensor layers_1_fc1_weight_to_fp16 = const()[name = string("layers_1_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(57665984)))]; tensor layers_1_fc1_bias_to_fp16 = const()[name = string("layers_1_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(66054656)))]; tensor x_31_cast_fp16 = conv(bias = layers_1_fc1_bias_to_fp16, dilations = x_31_dilations_0, groups = x_31_groups_0, pad = x_31_pad_0, pad_type = x_31_pad_type_0, strides = x_31_strides_0, weight = layers_1_fc1_weight_to_fp16, x = input_21_cast_fp16)[name = string("x_31_cast_fp16")]; fp16 var_764_to_fp16 = const()[name = string("op_764_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_765_cast_fp16 = mul(x = x_31_cast_fp16, y = var_764_to_fp16)[name = string("op_765_cast_fp16")]; tensor var_766_cast_fp16 = mul(x = var_765_cast_fp16, y = x_31_cast_fp16)[name = string("op_766_cast_fp16")]; tensor var_767_cast_fp16 = mul(x = var_766_cast_fp16, y = x_31_cast_fp16)[name = string("op_767_cast_fp16")]; tensor var_768_cast_fp16 = add(x = x_31_cast_fp16, y = var_767_cast_fp16)[name = string("op_768_cast_fp16")]; fp16 var_769_to_fp16 = const()[name = string("op_769_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_9_cast_fp16 = mul(x = var_768_cast_fp16, y = var_769_to_fp16)[name = string("u_9_cast_fp16")]; fp16 var_771_to_fp16 = const()[name = string("op_771_to_fp16"), val = fp16(0x1p-1)]; tensor var_772_cast_fp16 = mul(x = x_31_cast_fp16, y = var_771_to_fp16)[name = string("op_772_cast_fp16")]; tensor var_773_cast_fp16 = tanh(x = u_9_cast_fp16)[name = string("op_773_cast_fp16")]; fp16 var_774_to_fp16 = const()[name = string("op_774_to_fp16"), val = fp16(0x1p+0)]; tensor var_775_cast_fp16 = add(x = var_773_cast_fp16, y = var_774_to_fp16)[name = string("op_775_cast_fp16")]; tensor input_23_cast_fp16 = mul(x = var_772_cast_fp16, y = var_775_cast_fp16)[name = string("input_23_cast_fp16")]; string h_3_pad_type_0 = const()[name = string("h_3_pad_type_0"), val = string("valid")]; tensor h_3_strides_0 = const()[name = string("h_3_strides_0"), val = tensor([1, 1])]; tensor h_3_pad_0 = const()[name = string("h_3_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_3_dilations_0 = const()[name = string("h_3_dilations_0"), val = tensor([1, 1])]; int32 h_3_groups_0 = const()[name = string("h_3_groups_0"), val = int32(1)]; tensor layers_1_fc2_weight_to_fp16 = const()[name = string("layers_1_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(66062912)))]; tensor layers_1_fc2_bias_to_fp16 = const()[name = string("layers_1_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(74451584)))]; tensor h_3_cast_fp16 = conv(bias = layers_1_fc2_bias_to_fp16, dilations = h_3_dilations_0, groups = h_3_groups_0, pad = h_3_pad_0, pad_type = h_3_pad_type_0, strides = h_3_strides_0, weight = layers_1_fc2_weight_to_fp16, x = input_23_cast_fp16)[name = string("h_3_cast_fp16")]; tensor x_33_cast_fp16 = add(x = x_27_cast_fp16, y = h_3_cast_fp16)[name = string("x_33_cast_fp16")]; int32 var_791 = const()[name = string("op_791"), val = int32(1)]; tensor mu_9_axes_0 = const()[name = string("mu_9_axes_0"), val = tensor([1])]; bool mu_9_keep_dims_0 = const()[name = string("mu_9_keep_dims_0"), val = bool(true)]; tensor mu_9_cast_fp16 = reduce_mean(axes = mu_9_axes_0, keep_dims = mu_9_keep_dims_0, x = x_33_cast_fp16)[name = string("mu_9_cast_fp16")]; tensor var_805_cast_fp16 = sub(x = x_33_cast_fp16, y = mu_9_cast_fp16)[name = string("op_805_cast_fp16")]; fp16 var_794_promoted_to_fp16 = const()[name = string("op_794_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_806_cast_fp16 = pow(x = var_805_cast_fp16, y = var_794_promoted_to_fp16)[name = string("op_806_cast_fp16")]; tensor var_9_axes_0 = const()[name = string("var_9_axes_0"), val = tensor([1])]; bool var_9_keep_dims_0 = const()[name = string("var_9_keep_dims_0"), val = bool(true)]; tensor var_9_cast_fp16 = reduce_mean(axes = var_9_axes_0, keep_dims = var_9_keep_dims_0, x = var_806_cast_fp16)[name = string("var_9_cast_fp16")]; fp16 var_810_to_fp16 = const()[name = string("op_810_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_811_cast_fp16 = add(x = var_9_cast_fp16, y = var_810_to_fp16)[name = string("op_811_cast_fp16")]; fp32 var_812_epsilon_0 = const()[name = string("op_812_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_812_cast_fp16 = rsqrt(epsilon = var_812_epsilon_0, x = var_811_cast_fp16)[name = string("op_812_cast_fp16")]; tensor x_35_cast_fp16 = mul(x = var_805_cast_fp16, y = var_812_cast_fp16)[name = string("x_35_cast_fp16")]; tensor input_25_gamma_0_to_fp16 = const()[name = string("input_25_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(74453696)))]; tensor input_25_beta_0_to_fp16 = const()[name = string("input_25_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(74455808)))]; fp16 input_25_epsilon_0_to_fp16 = const()[name = string("input_25_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_25_cast_fp16 = batch_norm(beta = input_25_beta_0_to_fp16, epsilon = input_25_epsilon_0_to_fp16, gamma = input_25_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_35_cast_fp16)[name = string("input_25_cast_fp16")]; string var_830_pad_type_0 = const()[name = string("op_830_pad_type_0"), val = string("valid")]; tensor var_830_strides_0 = const()[name = string("op_830_strides_0"), val = tensor([1, 1])]; tensor var_830_pad_0 = const()[name = string("op_830_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_830_dilations_0 = const()[name = string("op_830_dilations_0"), val = tensor([1, 1])]; int32 var_830_groups_0 = const()[name = string("op_830_groups_0"), val = int32(1)]; tensor var_832_weight_0_to_fp16 = const()[name = string("op_832_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(74457920)))]; tensor var_832_bias_0_to_fp16 = const()[name = string("op_832_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(76555136)))]; tensor var_832_cast_fp16 = conv(bias = var_832_bias_0_to_fp16, dilations = var_830_dilations_0, groups = var_830_groups_0, pad = var_830_pad_0, pad_type = var_830_pad_type_0, strides = var_830_strides_0, weight = var_832_weight_0_to_fp16, x = input_25_cast_fp16)[name = string("op_832_cast_fp16")]; string var_839_pad_type_0 = const()[name = string("op_839_pad_type_0"), val = string("valid")]; tensor var_839_strides_0 = const()[name = string("op_839_strides_0"), val = tensor([1, 1])]; tensor var_839_pad_0 = const()[name = string("op_839_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_839_dilations_0 = const()[name = string("op_839_dilations_0"), val = tensor([1, 1])]; int32 var_839_groups_0 = const()[name = string("op_839_groups_0"), val = int32(1)]; tensor layers_2_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_2_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(76557248)))]; tensor layers_2_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_2_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(78654464)))]; tensor var_839_cast_fp16 = conv(bias = layers_2_self_attn_k_proj_bias_to_fp16, dilations = var_839_dilations_0, groups = var_839_groups_0, pad = var_839_pad_0, pad_type = var_839_pad_type_0, strides = var_839_strides_0, weight = layers_2_self_attn_k_proj_weight_to_fp16, x = input_25_cast_fp16)[name = string("op_839_cast_fp16")]; string var_846_pad_type_0 = const()[name = string("op_846_pad_type_0"), val = string("valid")]; tensor var_846_strides_0 = const()[name = string("op_846_strides_0"), val = tensor([1, 1])]; tensor var_846_pad_0 = const()[name = string("op_846_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_846_dilations_0 = const()[name = string("op_846_dilations_0"), val = tensor([1, 1])]; int32 var_846_groups_0 = const()[name = string("op_846_groups_0"), val = int32(1)]; tensor layers_2_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_2_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(78656576)))]; tensor layers_2_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_2_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(80753792)))]; tensor var_846_cast_fp16 = conv(bias = layers_2_self_attn_v_proj_bias_to_fp16, dilations = var_846_dilations_0, groups = var_846_groups_0, pad = var_846_pad_0, pad_type = var_846_pad_type_0, strides = var_846_strides_0, weight = layers_2_self_attn_v_proj_weight_to_fp16, x = input_25_cast_fp16)[name = string("op_846_cast_fp16")]; tensor tile_6 = const()[name = string("tile_6"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(80755904)))]; int32 var_847_axis_0 = const()[name = string("op_847_axis_0"), val = int32(1)]; tensor var_847_cast_fp16_0, tensor var_847_cast_fp16_1, tensor var_847_cast_fp16_2, tensor var_847_cast_fp16_3, tensor var_847_cast_fp16_4, tensor var_847_cast_fp16_5, tensor var_847_cast_fp16_6, tensor var_847_cast_fp16_7, tensor var_847_cast_fp16_8, tensor var_847_cast_fp16_9, tensor var_847_cast_fp16_10, tensor var_847_cast_fp16_11, tensor var_847_cast_fp16_12, tensor var_847_cast_fp16_13, tensor var_847_cast_fp16_14, tensor var_847_cast_fp16_15 = split(axis = var_847_axis_0, split_sizes = tile_6, x = var_832_cast_fp16)[name = string("op_847_cast_fp16")]; tensor tile_7 = const()[name = string("tile_7"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(80756032)))]; int32 var_864_axis_0 = const()[name = string("op_864_axis_0"), val = int32(1)]; tensor var_864_cast_fp16_0, tensor var_864_cast_fp16_1, tensor var_864_cast_fp16_2, tensor var_864_cast_fp16_3, tensor var_864_cast_fp16_4, tensor var_864_cast_fp16_5, tensor var_864_cast_fp16_6, tensor var_864_cast_fp16_7, tensor var_864_cast_fp16_8, tensor var_864_cast_fp16_9, tensor var_864_cast_fp16_10, tensor var_864_cast_fp16_11, tensor var_864_cast_fp16_12, tensor var_864_cast_fp16_13, tensor var_864_cast_fp16_14, tensor var_864_cast_fp16_15 = split(axis = var_864_axis_0, split_sizes = tile_7, x = var_839_cast_fp16)[name = string("op_864_cast_fp16")]; tensor tile_8 = const()[name = string("tile_8"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(80756160)))]; int32 var_881_axis_0 = const()[name = string("op_881_axis_0"), val = int32(1)]; tensor var_881_cast_fp16_0, tensor var_881_cast_fp16_1, tensor var_881_cast_fp16_2, tensor var_881_cast_fp16_3, tensor var_881_cast_fp16_4, tensor var_881_cast_fp16_5, tensor var_881_cast_fp16_6, tensor var_881_cast_fp16_7, tensor var_881_cast_fp16_8, tensor var_881_cast_fp16_9, tensor var_881_cast_fp16_10, tensor var_881_cast_fp16_11, tensor var_881_cast_fp16_12, tensor var_881_cast_fp16_13, tensor var_881_cast_fp16_14, tensor var_881_cast_fp16_15 = split(axis = var_881_axis_0, split_sizes = tile_8, x = var_846_cast_fp16)[name = string("op_881_cast_fp16")]; tensor transpose_64_perm_0 = const()[name = string("transpose_64_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_324 = const()[name = string("concat_324"), val = tensor([1, 104, 64])]; tensor transpose_64_cast_fp16 = transpose(perm = transpose_64_perm_0, x = var_847_cast_fp16_0)[name = string("transpose_4127")]; tensor reshape_96_cast_fp16 = reshape(shape = concat_324, x = transpose_64_cast_fp16)[name = string("reshape_96_cast_fp16")]; tensor transpose_65_perm_0 = const()[name = string("transpose_65_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_325 = const()[name = string("concat_325"), val = tensor([1, 64, 104])]; tensor transpose_65_cast_fp16 = transpose(perm = transpose_65_perm_0, x = var_864_cast_fp16_0)[name = string("transpose_4126")]; tensor reshape_97_cast_fp16 = reshape(shape = concat_325, x = transpose_65_cast_fp16)[name = string("reshape_97_cast_fp16")]; bool matmul_32_transpose_x_0 = const()[name = string("matmul_32_transpose_x_0"), val = bool(false)]; bool matmul_32_transpose_y_0 = const()[name = string("matmul_32_transpose_y_0"), val = bool(false)]; tensor matmul_32_cast_fp16 = matmul(transpose_x = matmul_32_transpose_x_0, transpose_y = matmul_32_transpose_y_0, x = reshape_96_cast_fp16, y = reshape_97_cast_fp16)[name = string("matmul_32_cast_fp16")]; tensor concat_329 = const()[name = string("concat_329"), val = tensor([1, 1, 104, 104])]; tensor reshape_98_cast_fp16 = reshape(shape = concat_329, x = matmul_32_cast_fp16)[name = string("reshape_98_cast_fp16")]; tensor transpose_2720_perm_0 = const()[name = string("transpose_2720_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2720 = transpose(perm = transpose_2720_perm_0, x = reshape_98_cast_fp16)[name = string("transpose_4125")]; tensor w_131_cast_fp16 = add(x = transpose_2720, y = transpose_2305)[name = string("w_131_cast_fp16")]; tensor var_903_cast_fp16 = softmax(axis = var_791, x = w_131_cast_fp16)[name = string("op_903_cast_fp16")]; string var_905_equation_0 = const()[name = string("op_905_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_905_cast_fp16 = einsum(equation = var_905_equation_0, values = (var_881_cast_fp16_0, var_903_cast_fp16))[name = string("op_905_cast_fp16")]; tensor transpose_66_perm_0 = const()[name = string("transpose_66_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_334 = const()[name = string("concat_334"), val = tensor([1, 104, 64])]; tensor transpose_66_cast_fp16 = transpose(perm = transpose_66_perm_0, x = var_847_cast_fp16_1)[name = string("transpose_4124")]; tensor reshape_99_cast_fp16 = reshape(shape = concat_334, x = transpose_66_cast_fp16)[name = string("reshape_99_cast_fp16")]; tensor transpose_67_perm_0 = const()[name = string("transpose_67_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_335 = const()[name = string("concat_335"), val = tensor([1, 64, 104])]; tensor transpose_67_cast_fp16 = transpose(perm = transpose_67_perm_0, x = var_864_cast_fp16_1)[name = string("transpose_4123")]; tensor reshape_100_cast_fp16 = reshape(shape = concat_335, x = transpose_67_cast_fp16)[name = string("reshape_100_cast_fp16")]; bool matmul_33_transpose_x_0 = const()[name = string("matmul_33_transpose_x_0"), val = bool(false)]; bool matmul_33_transpose_y_0 = const()[name = string("matmul_33_transpose_y_0"), val = bool(false)]; tensor matmul_33_cast_fp16 = matmul(transpose_x = matmul_33_transpose_x_0, transpose_y = matmul_33_transpose_y_0, x = reshape_99_cast_fp16, y = reshape_100_cast_fp16)[name = string("matmul_33_cast_fp16")]; tensor concat_339 = const()[name = string("concat_339"), val = tensor([1, 1, 104, 104])]; tensor reshape_101_cast_fp16 = reshape(shape = concat_339, x = matmul_33_cast_fp16)[name = string("reshape_101_cast_fp16")]; tensor transpose_2721_perm_0 = const()[name = string("transpose_2721_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2721 = transpose(perm = transpose_2721_perm_0, x = reshape_101_cast_fp16)[name = string("transpose_4122")]; tensor w_135_cast_fp16 = add(x = transpose_2721, y = transpose_2305)[name = string("w_135_cast_fp16")]; tensor var_911_cast_fp16 = softmax(axis = var_791, x = w_135_cast_fp16)[name = string("op_911_cast_fp16")]; string var_913_equation_0 = const()[name = string("op_913_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_913_cast_fp16 = einsum(equation = var_913_equation_0, values = (var_881_cast_fp16_1, var_911_cast_fp16))[name = string("op_913_cast_fp16")]; tensor transpose_68_perm_0 = const()[name = string("transpose_68_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_344 = const()[name = string("concat_344"), val = tensor([1, 104, 64])]; tensor transpose_68_cast_fp16 = transpose(perm = transpose_68_perm_0, x = var_847_cast_fp16_2)[name = string("transpose_4121")]; tensor reshape_102_cast_fp16 = reshape(shape = concat_344, x = transpose_68_cast_fp16)[name = string("reshape_102_cast_fp16")]; tensor transpose_69_perm_0 = const()[name = string("transpose_69_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_345 = const()[name = string("concat_345"), val = tensor([1, 64, 104])]; tensor transpose_69_cast_fp16 = transpose(perm = transpose_69_perm_0, x = var_864_cast_fp16_2)[name = string("transpose_4120")]; tensor reshape_103_cast_fp16 = reshape(shape = concat_345, x = transpose_69_cast_fp16)[name = string("reshape_103_cast_fp16")]; bool matmul_34_transpose_x_0 = const()[name = string("matmul_34_transpose_x_0"), val = bool(false)]; bool matmul_34_transpose_y_0 = const()[name = string("matmul_34_transpose_y_0"), val = bool(false)]; tensor matmul_34_cast_fp16 = matmul(transpose_x = matmul_34_transpose_x_0, transpose_y = matmul_34_transpose_y_0, x = reshape_102_cast_fp16, y = reshape_103_cast_fp16)[name = string("matmul_34_cast_fp16")]; tensor concat_349 = const()[name = string("concat_349"), val = tensor([1, 1, 104, 104])]; tensor reshape_104_cast_fp16 = reshape(shape = concat_349, x = matmul_34_cast_fp16)[name = string("reshape_104_cast_fp16")]; tensor transpose_2722_perm_0 = const()[name = string("transpose_2722_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2722 = transpose(perm = transpose_2722_perm_0, x = reshape_104_cast_fp16)[name = string("transpose_4119")]; tensor w_139_cast_fp16 = add(x = transpose_2722, y = transpose_2305)[name = string("w_139_cast_fp16")]; tensor var_919_cast_fp16 = softmax(axis = var_791, x = w_139_cast_fp16)[name = string("op_919_cast_fp16")]; string var_921_equation_0 = const()[name = string("op_921_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_921_cast_fp16 = einsum(equation = var_921_equation_0, values = (var_881_cast_fp16_2, var_919_cast_fp16))[name = string("op_921_cast_fp16")]; tensor transpose_70_perm_0 = const()[name = string("transpose_70_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_354 = const()[name = string("concat_354"), val = tensor([1, 104, 64])]; tensor transpose_70_cast_fp16 = transpose(perm = transpose_70_perm_0, x = var_847_cast_fp16_3)[name = string("transpose_4118")]; tensor reshape_105_cast_fp16 = reshape(shape = concat_354, x = transpose_70_cast_fp16)[name = string("reshape_105_cast_fp16")]; tensor transpose_71_perm_0 = const()[name = string("transpose_71_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_355 = const()[name = string("concat_355"), val = tensor([1, 64, 104])]; tensor transpose_71_cast_fp16 = transpose(perm = transpose_71_perm_0, x = var_864_cast_fp16_3)[name = string("transpose_4117")]; tensor reshape_106_cast_fp16 = reshape(shape = concat_355, x = transpose_71_cast_fp16)[name = string("reshape_106_cast_fp16")]; bool matmul_35_transpose_x_0 = const()[name = string("matmul_35_transpose_x_0"), val = bool(false)]; bool matmul_35_transpose_y_0 = const()[name = string("matmul_35_transpose_y_0"), val = bool(false)]; tensor matmul_35_cast_fp16 = matmul(transpose_x = matmul_35_transpose_x_0, transpose_y = matmul_35_transpose_y_0, x = reshape_105_cast_fp16, y = reshape_106_cast_fp16)[name = string("matmul_35_cast_fp16")]; tensor concat_359 = const()[name = string("concat_359"), val = tensor([1, 1, 104, 104])]; tensor reshape_107_cast_fp16 = reshape(shape = concat_359, x = matmul_35_cast_fp16)[name = string("reshape_107_cast_fp16")]; tensor transpose_2723_perm_0 = const()[name = string("transpose_2723_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2723 = transpose(perm = transpose_2723_perm_0, x = reshape_107_cast_fp16)[name = string("transpose_4116")]; tensor w_143_cast_fp16 = add(x = transpose_2723, y = transpose_2305)[name = string("w_143_cast_fp16")]; tensor var_927_cast_fp16 = softmax(axis = var_791, x = w_143_cast_fp16)[name = string("op_927_cast_fp16")]; string var_929_equation_0 = const()[name = string("op_929_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_929_cast_fp16 = einsum(equation = var_929_equation_0, values = (var_881_cast_fp16_3, var_927_cast_fp16))[name = string("op_929_cast_fp16")]; tensor transpose_72_perm_0 = const()[name = string("transpose_72_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_364 = const()[name = string("concat_364"), val = tensor([1, 104, 64])]; tensor transpose_72_cast_fp16 = transpose(perm = transpose_72_perm_0, x = var_847_cast_fp16_4)[name = string("transpose_4115")]; tensor reshape_108_cast_fp16 = reshape(shape = concat_364, x = transpose_72_cast_fp16)[name = string("reshape_108_cast_fp16")]; tensor transpose_73_perm_0 = const()[name = string("transpose_73_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_365 = const()[name = string("concat_365"), val = tensor([1, 64, 104])]; tensor transpose_73_cast_fp16 = transpose(perm = transpose_73_perm_0, x = var_864_cast_fp16_4)[name = string("transpose_4114")]; tensor reshape_109_cast_fp16 = reshape(shape = concat_365, x = transpose_73_cast_fp16)[name = string("reshape_109_cast_fp16")]; bool matmul_36_transpose_x_0 = const()[name = string("matmul_36_transpose_x_0"), val = bool(false)]; bool matmul_36_transpose_y_0 = const()[name = string("matmul_36_transpose_y_0"), val = bool(false)]; tensor matmul_36_cast_fp16 = matmul(transpose_x = matmul_36_transpose_x_0, transpose_y = matmul_36_transpose_y_0, x = reshape_108_cast_fp16, y = reshape_109_cast_fp16)[name = string("matmul_36_cast_fp16")]; tensor concat_369 = const()[name = string("concat_369"), val = tensor([1, 1, 104, 104])]; tensor reshape_110_cast_fp16 = reshape(shape = concat_369, x = matmul_36_cast_fp16)[name = string("reshape_110_cast_fp16")]; tensor transpose_2724_perm_0 = const()[name = string("transpose_2724_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2724 = transpose(perm = transpose_2724_perm_0, x = reshape_110_cast_fp16)[name = string("transpose_4113")]; tensor w_147_cast_fp16 = add(x = transpose_2724, y = transpose_2305)[name = string("w_147_cast_fp16")]; tensor var_935_cast_fp16 = softmax(axis = var_791, x = w_147_cast_fp16)[name = string("op_935_cast_fp16")]; string var_937_equation_0 = const()[name = string("op_937_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_937_cast_fp16 = einsum(equation = var_937_equation_0, values = (var_881_cast_fp16_4, var_935_cast_fp16))[name = string("op_937_cast_fp16")]; tensor transpose_74_perm_0 = const()[name = string("transpose_74_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_374 = const()[name = string("concat_374"), val = tensor([1, 104, 64])]; tensor transpose_74_cast_fp16 = transpose(perm = transpose_74_perm_0, x = var_847_cast_fp16_5)[name = string("transpose_4112")]; tensor reshape_111_cast_fp16 = reshape(shape = concat_374, x = transpose_74_cast_fp16)[name = string("reshape_111_cast_fp16")]; tensor transpose_75_perm_0 = const()[name = string("transpose_75_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_375 = const()[name = string("concat_375"), val = tensor([1, 64, 104])]; tensor transpose_75_cast_fp16 = transpose(perm = transpose_75_perm_0, x = var_864_cast_fp16_5)[name = string("transpose_4111")]; tensor reshape_112_cast_fp16 = reshape(shape = concat_375, x = transpose_75_cast_fp16)[name = string("reshape_112_cast_fp16")]; bool matmul_37_transpose_x_0 = const()[name = string("matmul_37_transpose_x_0"), val = bool(false)]; bool matmul_37_transpose_y_0 = const()[name = string("matmul_37_transpose_y_0"), val = bool(false)]; tensor matmul_37_cast_fp16 = matmul(transpose_x = matmul_37_transpose_x_0, transpose_y = matmul_37_transpose_y_0, x = reshape_111_cast_fp16, y = reshape_112_cast_fp16)[name = string("matmul_37_cast_fp16")]; tensor concat_379 = const()[name = string("concat_379"), val = tensor([1, 1, 104, 104])]; tensor reshape_113_cast_fp16 = reshape(shape = concat_379, x = matmul_37_cast_fp16)[name = string("reshape_113_cast_fp16")]; tensor transpose_2725_perm_0 = const()[name = string("transpose_2725_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2725 = transpose(perm = transpose_2725_perm_0, x = reshape_113_cast_fp16)[name = string("transpose_4110")]; tensor w_151_cast_fp16 = add(x = transpose_2725, y = transpose_2305)[name = string("w_151_cast_fp16")]; tensor var_943_cast_fp16 = softmax(axis = var_791, x = w_151_cast_fp16)[name = string("op_943_cast_fp16")]; string var_945_equation_0 = const()[name = string("op_945_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_945_cast_fp16 = einsum(equation = var_945_equation_0, values = (var_881_cast_fp16_5, var_943_cast_fp16))[name = string("op_945_cast_fp16")]; tensor transpose_76_perm_0 = const()[name = string("transpose_76_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_384 = const()[name = string("concat_384"), val = tensor([1, 104, 64])]; tensor transpose_76_cast_fp16 = transpose(perm = transpose_76_perm_0, x = var_847_cast_fp16_6)[name = string("transpose_4109")]; tensor reshape_114_cast_fp16 = reshape(shape = concat_384, x = transpose_76_cast_fp16)[name = string("reshape_114_cast_fp16")]; tensor transpose_77_perm_0 = const()[name = string("transpose_77_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_385 = const()[name = string("concat_385"), val = tensor([1, 64, 104])]; tensor transpose_77_cast_fp16 = transpose(perm = transpose_77_perm_0, x = var_864_cast_fp16_6)[name = string("transpose_4108")]; tensor reshape_115_cast_fp16 = reshape(shape = concat_385, x = transpose_77_cast_fp16)[name = string("reshape_115_cast_fp16")]; bool matmul_38_transpose_x_0 = const()[name = string("matmul_38_transpose_x_0"), val = bool(false)]; bool matmul_38_transpose_y_0 = const()[name = string("matmul_38_transpose_y_0"), val = bool(false)]; tensor matmul_38_cast_fp16 = matmul(transpose_x = matmul_38_transpose_x_0, transpose_y = matmul_38_transpose_y_0, x = reshape_114_cast_fp16, y = reshape_115_cast_fp16)[name = string("matmul_38_cast_fp16")]; tensor concat_389 = const()[name = string("concat_389"), val = tensor([1, 1, 104, 104])]; tensor reshape_116_cast_fp16 = reshape(shape = concat_389, x = matmul_38_cast_fp16)[name = string("reshape_116_cast_fp16")]; tensor transpose_2726_perm_0 = const()[name = string("transpose_2726_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2726 = transpose(perm = transpose_2726_perm_0, x = reshape_116_cast_fp16)[name = string("transpose_4107")]; tensor w_155_cast_fp16 = add(x = transpose_2726, y = transpose_2305)[name = string("w_155_cast_fp16")]; tensor var_951_cast_fp16 = softmax(axis = var_791, x = w_155_cast_fp16)[name = string("op_951_cast_fp16")]; string var_953_equation_0 = const()[name = string("op_953_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_953_cast_fp16 = einsum(equation = var_953_equation_0, values = (var_881_cast_fp16_6, var_951_cast_fp16))[name = string("op_953_cast_fp16")]; tensor transpose_78_perm_0 = const()[name = string("transpose_78_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_394 = const()[name = string("concat_394"), val = tensor([1, 104, 64])]; tensor transpose_78_cast_fp16 = transpose(perm = transpose_78_perm_0, x = var_847_cast_fp16_7)[name = string("transpose_4106")]; tensor reshape_117_cast_fp16 = reshape(shape = concat_394, x = transpose_78_cast_fp16)[name = string("reshape_117_cast_fp16")]; tensor transpose_79_perm_0 = const()[name = string("transpose_79_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_395 = const()[name = string("concat_395"), val = tensor([1, 64, 104])]; tensor transpose_79_cast_fp16 = transpose(perm = transpose_79_perm_0, x = var_864_cast_fp16_7)[name = string("transpose_4105")]; tensor reshape_118_cast_fp16 = reshape(shape = concat_395, x = transpose_79_cast_fp16)[name = string("reshape_118_cast_fp16")]; bool matmul_39_transpose_x_0 = const()[name = string("matmul_39_transpose_x_0"), val = bool(false)]; bool matmul_39_transpose_y_0 = const()[name = string("matmul_39_transpose_y_0"), val = bool(false)]; tensor matmul_39_cast_fp16 = matmul(transpose_x = matmul_39_transpose_x_0, transpose_y = matmul_39_transpose_y_0, x = reshape_117_cast_fp16, y = reshape_118_cast_fp16)[name = string("matmul_39_cast_fp16")]; tensor concat_399 = const()[name = string("concat_399"), val = tensor([1, 1, 104, 104])]; tensor reshape_119_cast_fp16 = reshape(shape = concat_399, x = matmul_39_cast_fp16)[name = string("reshape_119_cast_fp16")]; tensor transpose_2727_perm_0 = const()[name = string("transpose_2727_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2727 = transpose(perm = transpose_2727_perm_0, x = reshape_119_cast_fp16)[name = string("transpose_4104")]; tensor w_159_cast_fp16 = add(x = transpose_2727, y = transpose_2305)[name = string("w_159_cast_fp16")]; tensor var_959_cast_fp16 = softmax(axis = var_791, x = w_159_cast_fp16)[name = string("op_959_cast_fp16")]; string var_961_equation_0 = const()[name = string("op_961_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_961_cast_fp16 = einsum(equation = var_961_equation_0, values = (var_881_cast_fp16_7, var_959_cast_fp16))[name = string("op_961_cast_fp16")]; tensor transpose_80_perm_0 = const()[name = string("transpose_80_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_404 = const()[name = string("concat_404"), val = tensor([1, 104, 64])]; tensor transpose_80_cast_fp16 = transpose(perm = transpose_80_perm_0, x = var_847_cast_fp16_8)[name = string("transpose_4103")]; tensor reshape_120_cast_fp16 = reshape(shape = concat_404, x = transpose_80_cast_fp16)[name = string("reshape_120_cast_fp16")]; tensor transpose_81_perm_0 = const()[name = string("transpose_81_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_405 = const()[name = string("concat_405"), val = tensor([1, 64, 104])]; tensor transpose_81_cast_fp16 = transpose(perm = transpose_81_perm_0, x = var_864_cast_fp16_8)[name = string("transpose_4102")]; tensor reshape_121_cast_fp16 = reshape(shape = concat_405, x = transpose_81_cast_fp16)[name = string("reshape_121_cast_fp16")]; bool matmul_40_transpose_x_0 = const()[name = string("matmul_40_transpose_x_0"), val = bool(false)]; bool matmul_40_transpose_y_0 = const()[name = string("matmul_40_transpose_y_0"), val = bool(false)]; tensor matmul_40_cast_fp16 = matmul(transpose_x = matmul_40_transpose_x_0, transpose_y = matmul_40_transpose_y_0, x = reshape_120_cast_fp16, y = reshape_121_cast_fp16)[name = string("matmul_40_cast_fp16")]; tensor concat_409 = const()[name = string("concat_409"), val = tensor([1, 1, 104, 104])]; tensor reshape_122_cast_fp16 = reshape(shape = concat_409, x = matmul_40_cast_fp16)[name = string("reshape_122_cast_fp16")]; tensor transpose_2728_perm_0 = const()[name = string("transpose_2728_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2728 = transpose(perm = transpose_2728_perm_0, x = reshape_122_cast_fp16)[name = string("transpose_4101")]; tensor w_163_cast_fp16 = add(x = transpose_2728, y = transpose_2305)[name = string("w_163_cast_fp16")]; tensor var_967_cast_fp16 = softmax(axis = var_791, x = w_163_cast_fp16)[name = string("op_967_cast_fp16")]; string var_969_equation_0 = const()[name = string("op_969_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_969_cast_fp16 = einsum(equation = var_969_equation_0, values = (var_881_cast_fp16_8, var_967_cast_fp16))[name = string("op_969_cast_fp16")]; tensor transpose_82_perm_0 = const()[name = string("transpose_82_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_414 = const()[name = string("concat_414"), val = tensor([1, 104, 64])]; tensor transpose_82_cast_fp16 = transpose(perm = transpose_82_perm_0, x = var_847_cast_fp16_9)[name = string("transpose_4100")]; tensor reshape_123_cast_fp16 = reshape(shape = concat_414, x = transpose_82_cast_fp16)[name = string("reshape_123_cast_fp16")]; tensor transpose_83_perm_0 = const()[name = string("transpose_83_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_415 = const()[name = string("concat_415"), val = tensor([1, 64, 104])]; tensor transpose_83_cast_fp16 = transpose(perm = transpose_83_perm_0, x = var_864_cast_fp16_9)[name = string("transpose_4099")]; tensor reshape_124_cast_fp16 = reshape(shape = concat_415, x = transpose_83_cast_fp16)[name = string("reshape_124_cast_fp16")]; bool matmul_41_transpose_x_0 = const()[name = string("matmul_41_transpose_x_0"), val = bool(false)]; bool matmul_41_transpose_y_0 = const()[name = string("matmul_41_transpose_y_0"), val = bool(false)]; tensor matmul_41_cast_fp16 = matmul(transpose_x = matmul_41_transpose_x_0, transpose_y = matmul_41_transpose_y_0, x = reshape_123_cast_fp16, y = reshape_124_cast_fp16)[name = string("matmul_41_cast_fp16")]; tensor concat_419 = const()[name = string("concat_419"), val = tensor([1, 1, 104, 104])]; tensor reshape_125_cast_fp16 = reshape(shape = concat_419, x = matmul_41_cast_fp16)[name = string("reshape_125_cast_fp16")]; tensor transpose_2729_perm_0 = const()[name = string("transpose_2729_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2729 = transpose(perm = transpose_2729_perm_0, x = reshape_125_cast_fp16)[name = string("transpose_4098")]; tensor w_167_cast_fp16 = add(x = transpose_2729, y = transpose_2305)[name = string("w_167_cast_fp16")]; tensor var_975_cast_fp16 = softmax(axis = var_791, x = w_167_cast_fp16)[name = string("op_975_cast_fp16")]; string var_977_equation_0 = const()[name = string("op_977_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_977_cast_fp16 = einsum(equation = var_977_equation_0, values = (var_881_cast_fp16_9, var_975_cast_fp16))[name = string("op_977_cast_fp16")]; tensor transpose_84_perm_0 = const()[name = string("transpose_84_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_424 = const()[name = string("concat_424"), val = tensor([1, 104, 64])]; tensor transpose_84_cast_fp16 = transpose(perm = transpose_84_perm_0, x = var_847_cast_fp16_10)[name = string("transpose_4097")]; tensor reshape_126_cast_fp16 = reshape(shape = concat_424, x = transpose_84_cast_fp16)[name = string("reshape_126_cast_fp16")]; tensor transpose_85_perm_0 = const()[name = string("transpose_85_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_425 = const()[name = string("concat_425"), val = tensor([1, 64, 104])]; tensor transpose_85_cast_fp16 = transpose(perm = transpose_85_perm_0, x = var_864_cast_fp16_10)[name = string("transpose_4096")]; tensor reshape_127_cast_fp16 = reshape(shape = concat_425, x = transpose_85_cast_fp16)[name = string("reshape_127_cast_fp16")]; bool matmul_42_transpose_x_0 = const()[name = string("matmul_42_transpose_x_0"), val = bool(false)]; bool matmul_42_transpose_y_0 = const()[name = string("matmul_42_transpose_y_0"), val = bool(false)]; tensor matmul_42_cast_fp16 = matmul(transpose_x = matmul_42_transpose_x_0, transpose_y = matmul_42_transpose_y_0, x = reshape_126_cast_fp16, y = reshape_127_cast_fp16)[name = string("matmul_42_cast_fp16")]; tensor concat_429 = const()[name = string("concat_429"), val = tensor([1, 1, 104, 104])]; tensor reshape_128_cast_fp16 = reshape(shape = concat_429, x = matmul_42_cast_fp16)[name = string("reshape_128_cast_fp16")]; tensor transpose_2730_perm_0 = const()[name = string("transpose_2730_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2730 = transpose(perm = transpose_2730_perm_0, x = reshape_128_cast_fp16)[name = string("transpose_4095")]; tensor w_171_cast_fp16 = add(x = transpose_2730, y = transpose_2305)[name = string("w_171_cast_fp16")]; tensor var_983_cast_fp16 = softmax(axis = var_791, x = w_171_cast_fp16)[name = string("op_983_cast_fp16")]; string var_985_equation_0 = const()[name = string("op_985_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_985_cast_fp16 = einsum(equation = var_985_equation_0, values = (var_881_cast_fp16_10, var_983_cast_fp16))[name = string("op_985_cast_fp16")]; tensor transpose_86_perm_0 = const()[name = string("transpose_86_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_434 = const()[name = string("concat_434"), val = tensor([1, 104, 64])]; tensor transpose_86_cast_fp16 = transpose(perm = transpose_86_perm_0, x = var_847_cast_fp16_11)[name = string("transpose_4094")]; tensor reshape_129_cast_fp16 = reshape(shape = concat_434, x = transpose_86_cast_fp16)[name = string("reshape_129_cast_fp16")]; tensor transpose_87_perm_0 = const()[name = string("transpose_87_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_435 = const()[name = string("concat_435"), val = tensor([1, 64, 104])]; tensor transpose_87_cast_fp16 = transpose(perm = transpose_87_perm_0, x = var_864_cast_fp16_11)[name = string("transpose_4093")]; tensor reshape_130_cast_fp16 = reshape(shape = concat_435, x = transpose_87_cast_fp16)[name = string("reshape_130_cast_fp16")]; bool matmul_43_transpose_x_0 = const()[name = string("matmul_43_transpose_x_0"), val = bool(false)]; bool matmul_43_transpose_y_0 = const()[name = string("matmul_43_transpose_y_0"), val = bool(false)]; tensor matmul_43_cast_fp16 = matmul(transpose_x = matmul_43_transpose_x_0, transpose_y = matmul_43_transpose_y_0, x = reshape_129_cast_fp16, y = reshape_130_cast_fp16)[name = string("matmul_43_cast_fp16")]; tensor concat_439 = const()[name = string("concat_439"), val = tensor([1, 1, 104, 104])]; tensor reshape_131_cast_fp16 = reshape(shape = concat_439, x = matmul_43_cast_fp16)[name = string("reshape_131_cast_fp16")]; tensor transpose_2731_perm_0 = const()[name = string("transpose_2731_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2731 = transpose(perm = transpose_2731_perm_0, x = reshape_131_cast_fp16)[name = string("transpose_4092")]; tensor w_175_cast_fp16 = add(x = transpose_2731, y = transpose_2305)[name = string("w_175_cast_fp16")]; tensor var_991_cast_fp16 = softmax(axis = var_791, x = w_175_cast_fp16)[name = string("op_991_cast_fp16")]; string var_993_equation_0 = const()[name = string("op_993_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_993_cast_fp16 = einsum(equation = var_993_equation_0, values = (var_881_cast_fp16_11, var_991_cast_fp16))[name = string("op_993_cast_fp16")]; tensor transpose_88_perm_0 = const()[name = string("transpose_88_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_444 = const()[name = string("concat_444"), val = tensor([1, 104, 64])]; tensor transpose_88_cast_fp16 = transpose(perm = transpose_88_perm_0, x = var_847_cast_fp16_12)[name = string("transpose_4091")]; tensor reshape_132_cast_fp16 = reshape(shape = concat_444, x = transpose_88_cast_fp16)[name = string("reshape_132_cast_fp16")]; tensor transpose_89_perm_0 = const()[name = string("transpose_89_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_445 = const()[name = string("concat_445"), val = tensor([1, 64, 104])]; tensor transpose_89_cast_fp16 = transpose(perm = transpose_89_perm_0, x = var_864_cast_fp16_12)[name = string("transpose_4090")]; tensor reshape_133_cast_fp16 = reshape(shape = concat_445, x = transpose_89_cast_fp16)[name = string("reshape_133_cast_fp16")]; bool matmul_44_transpose_x_0 = const()[name = string("matmul_44_transpose_x_0"), val = bool(false)]; bool matmul_44_transpose_y_0 = const()[name = string("matmul_44_transpose_y_0"), val = bool(false)]; tensor matmul_44_cast_fp16 = matmul(transpose_x = matmul_44_transpose_x_0, transpose_y = matmul_44_transpose_y_0, x = reshape_132_cast_fp16, y = reshape_133_cast_fp16)[name = string("matmul_44_cast_fp16")]; tensor concat_449 = const()[name = string("concat_449"), val = tensor([1, 1, 104, 104])]; tensor reshape_134_cast_fp16 = reshape(shape = concat_449, x = matmul_44_cast_fp16)[name = string("reshape_134_cast_fp16")]; tensor transpose_2732_perm_0 = const()[name = string("transpose_2732_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2732 = transpose(perm = transpose_2732_perm_0, x = reshape_134_cast_fp16)[name = string("transpose_4089")]; tensor w_179_cast_fp16 = add(x = transpose_2732, y = transpose_2305)[name = string("w_179_cast_fp16")]; tensor var_999_cast_fp16 = softmax(axis = var_791, x = w_179_cast_fp16)[name = string("op_999_cast_fp16")]; string var_1001_equation_0 = const()[name = string("op_1001_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1001_cast_fp16 = einsum(equation = var_1001_equation_0, values = (var_881_cast_fp16_12, var_999_cast_fp16))[name = string("op_1001_cast_fp16")]; tensor transpose_90_perm_0 = const()[name = string("transpose_90_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_454 = const()[name = string("concat_454"), val = tensor([1, 104, 64])]; tensor transpose_90_cast_fp16 = transpose(perm = transpose_90_perm_0, x = var_847_cast_fp16_13)[name = string("transpose_4088")]; tensor reshape_135_cast_fp16 = reshape(shape = concat_454, x = transpose_90_cast_fp16)[name = string("reshape_135_cast_fp16")]; tensor transpose_91_perm_0 = const()[name = string("transpose_91_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_455 = const()[name = string("concat_455"), val = tensor([1, 64, 104])]; tensor transpose_91_cast_fp16 = transpose(perm = transpose_91_perm_0, x = var_864_cast_fp16_13)[name = string("transpose_4087")]; tensor reshape_136_cast_fp16 = reshape(shape = concat_455, x = transpose_91_cast_fp16)[name = string("reshape_136_cast_fp16")]; bool matmul_45_transpose_x_0 = const()[name = string("matmul_45_transpose_x_0"), val = bool(false)]; bool matmul_45_transpose_y_0 = const()[name = string("matmul_45_transpose_y_0"), val = bool(false)]; tensor matmul_45_cast_fp16 = matmul(transpose_x = matmul_45_transpose_x_0, transpose_y = matmul_45_transpose_y_0, x = reshape_135_cast_fp16, y = reshape_136_cast_fp16)[name = string("matmul_45_cast_fp16")]; tensor concat_459 = const()[name = string("concat_459"), val = tensor([1, 1, 104, 104])]; tensor reshape_137_cast_fp16 = reshape(shape = concat_459, x = matmul_45_cast_fp16)[name = string("reshape_137_cast_fp16")]; tensor transpose_2733_perm_0 = const()[name = string("transpose_2733_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2733 = transpose(perm = transpose_2733_perm_0, x = reshape_137_cast_fp16)[name = string("transpose_4086")]; tensor w_183_cast_fp16 = add(x = transpose_2733, y = transpose_2305)[name = string("w_183_cast_fp16")]; tensor var_1007_cast_fp16 = softmax(axis = var_791, x = w_183_cast_fp16)[name = string("op_1007_cast_fp16")]; string var_1009_equation_0 = const()[name = string("op_1009_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1009_cast_fp16 = einsum(equation = var_1009_equation_0, values = (var_881_cast_fp16_13, var_1007_cast_fp16))[name = string("op_1009_cast_fp16")]; tensor transpose_92_perm_0 = const()[name = string("transpose_92_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_464 = const()[name = string("concat_464"), val = tensor([1, 104, 64])]; tensor transpose_92_cast_fp16 = transpose(perm = transpose_92_perm_0, x = var_847_cast_fp16_14)[name = string("transpose_4085")]; tensor reshape_138_cast_fp16 = reshape(shape = concat_464, x = transpose_92_cast_fp16)[name = string("reshape_138_cast_fp16")]; tensor transpose_93_perm_0 = const()[name = string("transpose_93_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_465 = const()[name = string("concat_465"), val = tensor([1, 64, 104])]; tensor transpose_93_cast_fp16 = transpose(perm = transpose_93_perm_0, x = var_864_cast_fp16_14)[name = string("transpose_4084")]; tensor reshape_139_cast_fp16 = reshape(shape = concat_465, x = transpose_93_cast_fp16)[name = string("reshape_139_cast_fp16")]; bool matmul_46_transpose_x_0 = const()[name = string("matmul_46_transpose_x_0"), val = bool(false)]; bool matmul_46_transpose_y_0 = const()[name = string("matmul_46_transpose_y_0"), val = bool(false)]; tensor matmul_46_cast_fp16 = matmul(transpose_x = matmul_46_transpose_x_0, transpose_y = matmul_46_transpose_y_0, x = reshape_138_cast_fp16, y = reshape_139_cast_fp16)[name = string("matmul_46_cast_fp16")]; tensor concat_469 = const()[name = string("concat_469"), val = tensor([1, 1, 104, 104])]; tensor reshape_140_cast_fp16 = reshape(shape = concat_469, x = matmul_46_cast_fp16)[name = string("reshape_140_cast_fp16")]; tensor transpose_2734_perm_0 = const()[name = string("transpose_2734_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2734 = transpose(perm = transpose_2734_perm_0, x = reshape_140_cast_fp16)[name = string("transpose_4083")]; tensor w_187_cast_fp16 = add(x = transpose_2734, y = transpose_2305)[name = string("w_187_cast_fp16")]; tensor var_1015_cast_fp16 = softmax(axis = var_791, x = w_187_cast_fp16)[name = string("op_1015_cast_fp16")]; string var_1017_equation_0 = const()[name = string("op_1017_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1017_cast_fp16 = einsum(equation = var_1017_equation_0, values = (var_881_cast_fp16_14, var_1015_cast_fp16))[name = string("op_1017_cast_fp16")]; tensor transpose_94_perm_0 = const()[name = string("transpose_94_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_474 = const()[name = string("concat_474"), val = tensor([1, 104, 64])]; tensor transpose_94_cast_fp16 = transpose(perm = transpose_94_perm_0, x = var_847_cast_fp16_15)[name = string("transpose_4082")]; tensor reshape_141_cast_fp16 = reshape(shape = concat_474, x = transpose_94_cast_fp16)[name = string("reshape_141_cast_fp16")]; tensor transpose_95_perm_0 = const()[name = string("transpose_95_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_475 = const()[name = string("concat_475"), val = tensor([1, 64, 104])]; tensor transpose_95_cast_fp16 = transpose(perm = transpose_95_perm_0, x = var_864_cast_fp16_15)[name = string("transpose_4081")]; tensor reshape_142_cast_fp16 = reshape(shape = concat_475, x = transpose_95_cast_fp16)[name = string("reshape_142_cast_fp16")]; bool matmul_47_transpose_x_0 = const()[name = string("matmul_47_transpose_x_0"), val = bool(false)]; bool matmul_47_transpose_y_0 = const()[name = string("matmul_47_transpose_y_0"), val = bool(false)]; tensor matmul_47_cast_fp16 = matmul(transpose_x = matmul_47_transpose_x_0, transpose_y = matmul_47_transpose_y_0, x = reshape_141_cast_fp16, y = reshape_142_cast_fp16)[name = string("matmul_47_cast_fp16")]; tensor concat_479 = const()[name = string("concat_479"), val = tensor([1, 1, 104, 104])]; tensor reshape_143_cast_fp16 = reshape(shape = concat_479, x = matmul_47_cast_fp16)[name = string("reshape_143_cast_fp16")]; tensor transpose_2735_perm_0 = const()[name = string("transpose_2735_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2735 = transpose(perm = transpose_2735_perm_0, x = reshape_143_cast_fp16)[name = string("transpose_4080")]; tensor w_191_cast_fp16 = add(x = transpose_2735, y = transpose_2305)[name = string("w_191_cast_fp16")]; tensor var_1023_cast_fp16 = softmax(axis = var_791, x = w_191_cast_fp16)[name = string("op_1023_cast_fp16")]; string var_1025_equation_0 = const()[name = string("op_1025_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1025_cast_fp16 = einsum(equation = var_1025_equation_0, values = (var_881_cast_fp16_15, var_1023_cast_fp16))[name = string("op_1025_cast_fp16")]; bool input_27_interleave_0 = const()[name = string("input_27_interleave_0"), val = bool(false)]; tensor input_27_cast_fp16 = concat(axis = var_791, interleave = input_27_interleave_0, values = (var_905_cast_fp16, var_913_cast_fp16, var_921_cast_fp16, var_929_cast_fp16, var_937_cast_fp16, var_945_cast_fp16, var_953_cast_fp16, var_961_cast_fp16, var_969_cast_fp16, var_977_cast_fp16, var_985_cast_fp16, var_993_cast_fp16, var_1001_cast_fp16, var_1009_cast_fp16, var_1017_cast_fp16, var_1025_cast_fp16))[name = string("input_27_cast_fp16")]; string var_1034_pad_type_0 = const()[name = string("op_1034_pad_type_0"), val = string("valid")]; tensor var_1034_strides_0 = const()[name = string("op_1034_strides_0"), val = tensor([1, 1])]; tensor var_1034_pad_0 = const()[name = string("op_1034_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1034_dilations_0 = const()[name = string("op_1034_dilations_0"), val = tensor([1, 1])]; int32 var_1034_groups_0 = const()[name = string("op_1034_groups_0"), val = int32(1)]; tensor layers_2_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_2_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(80756288)))]; tensor layers_2_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_2_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(82853504)))]; tensor var_1034_cast_fp16 = conv(bias = layers_2_self_attn_out_proj_bias_to_fp16, dilations = var_1034_dilations_0, groups = var_1034_groups_0, pad = var_1034_pad_0, pad_type = var_1034_pad_type_0, strides = var_1034_strides_0, weight = layers_2_self_attn_out_proj_weight_to_fp16, x = input_27_cast_fp16)[name = string("op_1034_cast_fp16")]; tensor x_37_cast_fp16 = add(x = x_33_cast_fp16, y = var_1034_cast_fp16)[name = string("x_37_cast_fp16")]; tensor mu_11_axes_0 = const()[name = string("mu_11_axes_0"), val = tensor([1])]; bool mu_11_keep_dims_0 = const()[name = string("mu_11_keep_dims_0"), val = bool(true)]; tensor mu_11_cast_fp16 = reduce_mean(axes = mu_11_axes_0, keep_dims = mu_11_keep_dims_0, x = x_37_cast_fp16)[name = string("mu_11_cast_fp16")]; tensor var_1040_cast_fp16 = sub(x = x_37_cast_fp16, y = mu_11_cast_fp16)[name = string("op_1040_cast_fp16")]; fp16 var_794_promoted_1_to_fp16 = const()[name = string("op_794_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_1041_cast_fp16 = pow(x = var_1040_cast_fp16, y = var_794_promoted_1_to_fp16)[name = string("op_1041_cast_fp16")]; tensor var_11_axes_0 = const()[name = string("var_11_axes_0"), val = tensor([1])]; bool var_11_keep_dims_0 = const()[name = string("var_11_keep_dims_0"), val = bool(true)]; tensor var_11_cast_fp16 = reduce_mean(axes = var_11_axes_0, keep_dims = var_11_keep_dims_0, x = var_1041_cast_fp16)[name = string("var_11_cast_fp16")]; fp16 var_1045_to_fp16 = const()[name = string("op_1045_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1046_cast_fp16 = add(x = var_11_cast_fp16, y = var_1045_to_fp16)[name = string("op_1046_cast_fp16")]; fp32 var_1047_epsilon_0 = const()[name = string("op_1047_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1047_cast_fp16 = rsqrt(epsilon = var_1047_epsilon_0, x = var_1046_cast_fp16)[name = string("op_1047_cast_fp16")]; tensor x_39_cast_fp16 = mul(x = var_1040_cast_fp16, y = var_1047_cast_fp16)[name = string("x_39_cast_fp16")]; tensor input_29_gamma_0_to_fp16 = const()[name = string("input_29_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(82855616)))]; tensor input_29_beta_0_to_fp16 = const()[name = string("input_29_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(82857728)))]; fp16 input_29_epsilon_0_to_fp16 = const()[name = string("input_29_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_29_cast_fp16 = batch_norm(beta = input_29_beta_0_to_fp16, epsilon = input_29_epsilon_0_to_fp16, gamma = input_29_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_39_cast_fp16)[name = string("input_29_cast_fp16")]; string x_41_pad_type_0 = const()[name = string("x_41_pad_type_0"), val = string("valid")]; tensor x_41_strides_0 = const()[name = string("x_41_strides_0"), val = tensor([1, 1])]; tensor x_41_pad_0 = const()[name = string("x_41_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_41_dilations_0 = const()[name = string("x_41_dilations_0"), val = tensor([1, 1])]; int32 x_41_groups_0 = const()[name = string("x_41_groups_0"), val = int32(1)]; tensor layers_2_fc1_weight_to_fp16 = const()[name = string("layers_2_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(82859840)))]; tensor layers_2_fc1_bias_to_fp16 = const()[name = string("layers_2_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(91248512)))]; tensor x_41_cast_fp16 = conv(bias = layers_2_fc1_bias_to_fp16, dilations = x_41_dilations_0, groups = x_41_groups_0, pad = x_41_pad_0, pad_type = x_41_pad_type_0, strides = x_41_strides_0, weight = layers_2_fc1_weight_to_fp16, x = input_29_cast_fp16)[name = string("x_41_cast_fp16")]; fp16 var_1062_to_fp16 = const()[name = string("op_1062_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_1063_cast_fp16 = mul(x = x_41_cast_fp16, y = var_1062_to_fp16)[name = string("op_1063_cast_fp16")]; tensor var_1064_cast_fp16 = mul(x = var_1063_cast_fp16, y = x_41_cast_fp16)[name = string("op_1064_cast_fp16")]; tensor var_1065_cast_fp16 = mul(x = var_1064_cast_fp16, y = x_41_cast_fp16)[name = string("op_1065_cast_fp16")]; tensor var_1066_cast_fp16 = add(x = x_41_cast_fp16, y = var_1065_cast_fp16)[name = string("op_1066_cast_fp16")]; fp16 var_1067_to_fp16 = const()[name = string("op_1067_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_11_cast_fp16 = mul(x = var_1066_cast_fp16, y = var_1067_to_fp16)[name = string("u_11_cast_fp16")]; fp16 var_1069_to_fp16 = const()[name = string("op_1069_to_fp16"), val = fp16(0x1p-1)]; tensor var_1070_cast_fp16 = mul(x = x_41_cast_fp16, y = var_1069_to_fp16)[name = string("op_1070_cast_fp16")]; tensor var_1071_cast_fp16 = tanh(x = u_11_cast_fp16)[name = string("op_1071_cast_fp16")]; fp16 var_1072_to_fp16 = const()[name = string("op_1072_to_fp16"), val = fp16(0x1p+0)]; tensor var_1073_cast_fp16 = add(x = var_1071_cast_fp16, y = var_1072_to_fp16)[name = string("op_1073_cast_fp16")]; tensor input_31_cast_fp16 = mul(x = var_1070_cast_fp16, y = var_1073_cast_fp16)[name = string("input_31_cast_fp16")]; string h_5_pad_type_0 = const()[name = string("h_5_pad_type_0"), val = string("valid")]; tensor h_5_strides_0 = const()[name = string("h_5_strides_0"), val = tensor([1, 1])]; tensor h_5_pad_0 = const()[name = string("h_5_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_5_dilations_0 = const()[name = string("h_5_dilations_0"), val = tensor([1, 1])]; int32 h_5_groups_0 = const()[name = string("h_5_groups_0"), val = int32(1)]; tensor layers_2_fc2_weight_to_fp16 = const()[name = string("layers_2_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(91256768)))]; tensor layers_2_fc2_bias_to_fp16 = const()[name = string("layers_2_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(99645440)))]; tensor h_5_cast_fp16 = conv(bias = layers_2_fc2_bias_to_fp16, dilations = h_5_dilations_0, groups = h_5_groups_0, pad = h_5_pad_0, pad_type = h_5_pad_type_0, strides = h_5_strides_0, weight = layers_2_fc2_weight_to_fp16, x = input_31_cast_fp16)[name = string("h_5_cast_fp16")]; tensor x_43_cast_fp16 = add(x = x_37_cast_fp16, y = h_5_cast_fp16)[name = string("x_43_cast_fp16")]; int32 var_1089 = const()[name = string("op_1089"), val = int32(1)]; tensor mu_13_axes_0 = const()[name = string("mu_13_axes_0"), val = tensor([1])]; bool mu_13_keep_dims_0 = const()[name = string("mu_13_keep_dims_0"), val = bool(true)]; tensor mu_13_cast_fp16 = reduce_mean(axes = mu_13_axes_0, keep_dims = mu_13_keep_dims_0, x = x_43_cast_fp16)[name = string("mu_13_cast_fp16")]; tensor var_1103_cast_fp16 = sub(x = x_43_cast_fp16, y = mu_13_cast_fp16)[name = string("op_1103_cast_fp16")]; fp16 var_1092_promoted_to_fp16 = const()[name = string("op_1092_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_1104_cast_fp16 = pow(x = var_1103_cast_fp16, y = var_1092_promoted_to_fp16)[name = string("op_1104_cast_fp16")]; tensor var_13_axes_0 = const()[name = string("var_13_axes_0"), val = tensor([1])]; bool var_13_keep_dims_0 = const()[name = string("var_13_keep_dims_0"), val = bool(true)]; tensor var_13_cast_fp16 = reduce_mean(axes = var_13_axes_0, keep_dims = var_13_keep_dims_0, x = var_1104_cast_fp16)[name = string("var_13_cast_fp16")]; fp16 var_1108_to_fp16 = const()[name = string("op_1108_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1109_cast_fp16 = add(x = var_13_cast_fp16, y = var_1108_to_fp16)[name = string("op_1109_cast_fp16")]; fp32 var_1110_epsilon_0 = const()[name = string("op_1110_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1110_cast_fp16 = rsqrt(epsilon = var_1110_epsilon_0, x = var_1109_cast_fp16)[name = string("op_1110_cast_fp16")]; tensor x_45_cast_fp16 = mul(x = var_1103_cast_fp16, y = var_1110_cast_fp16)[name = string("x_45_cast_fp16")]; tensor input_33_gamma_0_to_fp16 = const()[name = string("input_33_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(99647552)))]; tensor input_33_beta_0_to_fp16 = const()[name = string("input_33_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(99649664)))]; fp16 input_33_epsilon_0_to_fp16 = const()[name = string("input_33_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_33_cast_fp16 = batch_norm(beta = input_33_beta_0_to_fp16, epsilon = input_33_epsilon_0_to_fp16, gamma = input_33_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_45_cast_fp16)[name = string("input_33_cast_fp16")]; string var_1128_pad_type_0 = const()[name = string("op_1128_pad_type_0"), val = string("valid")]; tensor var_1128_strides_0 = const()[name = string("op_1128_strides_0"), val = tensor([1, 1])]; tensor var_1128_pad_0 = const()[name = string("op_1128_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1128_dilations_0 = const()[name = string("op_1128_dilations_0"), val = tensor([1, 1])]; int32 var_1128_groups_0 = const()[name = string("op_1128_groups_0"), val = int32(1)]; tensor var_1130_weight_0_to_fp16 = const()[name = string("op_1130_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(99651776)))]; tensor var_1130_bias_0_to_fp16 = const()[name = string("op_1130_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(101748992)))]; tensor var_1130_cast_fp16 = conv(bias = var_1130_bias_0_to_fp16, dilations = var_1128_dilations_0, groups = var_1128_groups_0, pad = var_1128_pad_0, pad_type = var_1128_pad_type_0, strides = var_1128_strides_0, weight = var_1130_weight_0_to_fp16, x = input_33_cast_fp16)[name = string("op_1130_cast_fp16")]; string var_1137_pad_type_0 = const()[name = string("op_1137_pad_type_0"), val = string("valid")]; tensor var_1137_strides_0 = const()[name = string("op_1137_strides_0"), val = tensor([1, 1])]; tensor var_1137_pad_0 = const()[name = string("op_1137_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1137_dilations_0 = const()[name = string("op_1137_dilations_0"), val = tensor([1, 1])]; int32 var_1137_groups_0 = const()[name = string("op_1137_groups_0"), val = int32(1)]; tensor layers_3_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_3_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(101751104)))]; tensor layers_3_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_3_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(103848320)))]; tensor var_1137_cast_fp16 = conv(bias = layers_3_self_attn_k_proj_bias_to_fp16, dilations = var_1137_dilations_0, groups = var_1137_groups_0, pad = var_1137_pad_0, pad_type = var_1137_pad_type_0, strides = var_1137_strides_0, weight = layers_3_self_attn_k_proj_weight_to_fp16, x = input_33_cast_fp16)[name = string("op_1137_cast_fp16")]; string var_1144_pad_type_0 = const()[name = string("op_1144_pad_type_0"), val = string("valid")]; tensor var_1144_strides_0 = const()[name = string("op_1144_strides_0"), val = tensor([1, 1])]; tensor var_1144_pad_0 = const()[name = string("op_1144_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1144_dilations_0 = const()[name = string("op_1144_dilations_0"), val = tensor([1, 1])]; int32 var_1144_groups_0 = const()[name = string("op_1144_groups_0"), val = int32(1)]; tensor layers_3_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_3_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(103850432)))]; tensor layers_3_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_3_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(105947648)))]; tensor var_1144_cast_fp16 = conv(bias = layers_3_self_attn_v_proj_bias_to_fp16, dilations = var_1144_dilations_0, groups = var_1144_groups_0, pad = var_1144_pad_0, pad_type = var_1144_pad_type_0, strides = var_1144_strides_0, weight = layers_3_self_attn_v_proj_weight_to_fp16, x = input_33_cast_fp16)[name = string("op_1144_cast_fp16")]; tensor tile_9 = const()[name = string("tile_9"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(105949760)))]; int32 var_1145_axis_0 = const()[name = string("op_1145_axis_0"), val = int32(1)]; tensor var_1145_cast_fp16_0, tensor var_1145_cast_fp16_1, tensor var_1145_cast_fp16_2, tensor var_1145_cast_fp16_3, tensor var_1145_cast_fp16_4, tensor var_1145_cast_fp16_5, tensor var_1145_cast_fp16_6, tensor var_1145_cast_fp16_7, tensor var_1145_cast_fp16_8, tensor var_1145_cast_fp16_9, tensor var_1145_cast_fp16_10, tensor var_1145_cast_fp16_11, tensor var_1145_cast_fp16_12, tensor var_1145_cast_fp16_13, tensor var_1145_cast_fp16_14, tensor var_1145_cast_fp16_15 = split(axis = var_1145_axis_0, split_sizes = tile_9, x = var_1130_cast_fp16)[name = string("op_1145_cast_fp16")]; tensor tile_10 = const()[name = string("tile_10"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(105949888)))]; int32 var_1162_axis_0 = const()[name = string("op_1162_axis_0"), val = int32(1)]; tensor var_1162_cast_fp16_0, tensor var_1162_cast_fp16_1, tensor var_1162_cast_fp16_2, tensor var_1162_cast_fp16_3, tensor var_1162_cast_fp16_4, tensor var_1162_cast_fp16_5, tensor var_1162_cast_fp16_6, tensor var_1162_cast_fp16_7, tensor var_1162_cast_fp16_8, tensor var_1162_cast_fp16_9, tensor var_1162_cast_fp16_10, tensor var_1162_cast_fp16_11, tensor var_1162_cast_fp16_12, tensor var_1162_cast_fp16_13, tensor var_1162_cast_fp16_14, tensor var_1162_cast_fp16_15 = split(axis = var_1162_axis_0, split_sizes = tile_10, x = var_1137_cast_fp16)[name = string("op_1162_cast_fp16")]; tensor tile_11 = const()[name = string("tile_11"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(105950016)))]; int32 var_1179_axis_0 = const()[name = string("op_1179_axis_0"), val = int32(1)]; tensor var_1179_cast_fp16_0, tensor var_1179_cast_fp16_1, tensor var_1179_cast_fp16_2, tensor var_1179_cast_fp16_3, tensor var_1179_cast_fp16_4, tensor var_1179_cast_fp16_5, tensor var_1179_cast_fp16_6, tensor var_1179_cast_fp16_7, tensor var_1179_cast_fp16_8, tensor var_1179_cast_fp16_9, tensor var_1179_cast_fp16_10, tensor var_1179_cast_fp16_11, tensor var_1179_cast_fp16_12, tensor var_1179_cast_fp16_13, tensor var_1179_cast_fp16_14, tensor var_1179_cast_fp16_15 = split(axis = var_1179_axis_0, split_sizes = tile_11, x = var_1144_cast_fp16)[name = string("op_1179_cast_fp16")]; tensor transpose_96_perm_0 = const()[name = string("transpose_96_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_484 = const()[name = string("concat_484"), val = tensor([1, 104, 64])]; tensor transpose_96_cast_fp16 = transpose(perm = transpose_96_perm_0, x = var_1145_cast_fp16_0)[name = string("transpose_4079")]; tensor reshape_144_cast_fp16 = reshape(shape = concat_484, x = transpose_96_cast_fp16)[name = string("reshape_144_cast_fp16")]; tensor transpose_97_perm_0 = const()[name = string("transpose_97_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_485 = const()[name = string("concat_485"), val = tensor([1, 64, 104])]; tensor transpose_97_cast_fp16 = transpose(perm = transpose_97_perm_0, x = var_1162_cast_fp16_0)[name = string("transpose_4078")]; tensor reshape_145_cast_fp16 = reshape(shape = concat_485, x = transpose_97_cast_fp16)[name = string("reshape_145_cast_fp16")]; bool matmul_48_transpose_x_0 = const()[name = string("matmul_48_transpose_x_0"), val = bool(false)]; bool matmul_48_transpose_y_0 = const()[name = string("matmul_48_transpose_y_0"), val = bool(false)]; tensor matmul_48_cast_fp16 = matmul(transpose_x = matmul_48_transpose_x_0, transpose_y = matmul_48_transpose_y_0, x = reshape_144_cast_fp16, y = reshape_145_cast_fp16)[name = string("matmul_48_cast_fp16")]; tensor concat_489 = const()[name = string("concat_489"), val = tensor([1, 1, 104, 104])]; tensor reshape_146_cast_fp16 = reshape(shape = concat_489, x = matmul_48_cast_fp16)[name = string("reshape_146_cast_fp16")]; tensor transpose_2736_perm_0 = const()[name = string("transpose_2736_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2736 = transpose(perm = transpose_2736_perm_0, x = reshape_146_cast_fp16)[name = string("transpose_4077")]; tensor w_195_cast_fp16 = add(x = transpose_2736, y = transpose_2305)[name = string("w_195_cast_fp16")]; tensor var_1201_cast_fp16 = softmax(axis = var_1089, x = w_195_cast_fp16)[name = string("op_1201_cast_fp16")]; string var_1203_equation_0 = const()[name = string("op_1203_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1203_cast_fp16 = einsum(equation = var_1203_equation_0, values = (var_1179_cast_fp16_0, var_1201_cast_fp16))[name = string("op_1203_cast_fp16")]; tensor transpose_98_perm_0 = const()[name = string("transpose_98_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_494 = const()[name = string("concat_494"), val = tensor([1, 104, 64])]; tensor transpose_98_cast_fp16 = transpose(perm = transpose_98_perm_0, x = var_1145_cast_fp16_1)[name = string("transpose_4076")]; tensor reshape_147_cast_fp16 = reshape(shape = concat_494, x = transpose_98_cast_fp16)[name = string("reshape_147_cast_fp16")]; tensor transpose_99_perm_0 = const()[name = string("transpose_99_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_495 = const()[name = string("concat_495"), val = tensor([1, 64, 104])]; tensor transpose_99_cast_fp16 = transpose(perm = transpose_99_perm_0, x = var_1162_cast_fp16_1)[name = string("transpose_4075")]; tensor reshape_148_cast_fp16 = reshape(shape = concat_495, x = transpose_99_cast_fp16)[name = string("reshape_148_cast_fp16")]; bool matmul_49_transpose_x_0 = const()[name = string("matmul_49_transpose_x_0"), val = bool(false)]; bool matmul_49_transpose_y_0 = const()[name = string("matmul_49_transpose_y_0"), val = bool(false)]; tensor matmul_49_cast_fp16 = matmul(transpose_x = matmul_49_transpose_x_0, transpose_y = matmul_49_transpose_y_0, x = reshape_147_cast_fp16, y = reshape_148_cast_fp16)[name = string("matmul_49_cast_fp16")]; tensor concat_499 = const()[name = string("concat_499"), val = tensor([1, 1, 104, 104])]; tensor reshape_149_cast_fp16 = reshape(shape = concat_499, x = matmul_49_cast_fp16)[name = string("reshape_149_cast_fp16")]; tensor transpose_2737_perm_0 = const()[name = string("transpose_2737_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2737 = transpose(perm = transpose_2737_perm_0, x = reshape_149_cast_fp16)[name = string("transpose_4074")]; tensor w_199_cast_fp16 = add(x = transpose_2737, y = transpose_2305)[name = string("w_199_cast_fp16")]; tensor var_1209_cast_fp16 = softmax(axis = var_1089, x = w_199_cast_fp16)[name = string("op_1209_cast_fp16")]; string var_1211_equation_0 = const()[name = string("op_1211_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1211_cast_fp16 = einsum(equation = var_1211_equation_0, values = (var_1179_cast_fp16_1, var_1209_cast_fp16))[name = string("op_1211_cast_fp16")]; tensor transpose_100_perm_0 = const()[name = string("transpose_100_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_504 = const()[name = string("concat_504"), val = tensor([1, 104, 64])]; tensor transpose_100_cast_fp16 = transpose(perm = transpose_100_perm_0, x = var_1145_cast_fp16_2)[name = string("transpose_4073")]; tensor reshape_150_cast_fp16 = reshape(shape = concat_504, x = transpose_100_cast_fp16)[name = string("reshape_150_cast_fp16")]; tensor transpose_101_perm_0 = const()[name = string("transpose_101_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_505 = const()[name = string("concat_505"), val = tensor([1, 64, 104])]; tensor transpose_101_cast_fp16 = transpose(perm = transpose_101_perm_0, x = var_1162_cast_fp16_2)[name = string("transpose_4072")]; tensor reshape_151_cast_fp16 = reshape(shape = concat_505, x = transpose_101_cast_fp16)[name = string("reshape_151_cast_fp16")]; bool matmul_50_transpose_x_0 = const()[name = string("matmul_50_transpose_x_0"), val = bool(false)]; bool matmul_50_transpose_y_0 = const()[name = string("matmul_50_transpose_y_0"), val = bool(false)]; tensor matmul_50_cast_fp16 = matmul(transpose_x = matmul_50_transpose_x_0, transpose_y = matmul_50_transpose_y_0, x = reshape_150_cast_fp16, y = reshape_151_cast_fp16)[name = string("matmul_50_cast_fp16")]; tensor concat_509 = const()[name = string("concat_509"), val = tensor([1, 1, 104, 104])]; tensor reshape_152_cast_fp16 = reshape(shape = concat_509, x = matmul_50_cast_fp16)[name = string("reshape_152_cast_fp16")]; tensor transpose_2738_perm_0 = const()[name = string("transpose_2738_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2738 = transpose(perm = transpose_2738_perm_0, x = reshape_152_cast_fp16)[name = string("transpose_4071")]; tensor w_203_cast_fp16 = add(x = transpose_2738, y = transpose_2305)[name = string("w_203_cast_fp16")]; tensor var_1217_cast_fp16 = softmax(axis = var_1089, x = w_203_cast_fp16)[name = string("op_1217_cast_fp16")]; string var_1219_equation_0 = const()[name = string("op_1219_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1219_cast_fp16 = einsum(equation = var_1219_equation_0, values = (var_1179_cast_fp16_2, var_1217_cast_fp16))[name = string("op_1219_cast_fp16")]; tensor transpose_102_perm_0 = const()[name = string("transpose_102_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_514 = const()[name = string("concat_514"), val = tensor([1, 104, 64])]; tensor transpose_102_cast_fp16 = transpose(perm = transpose_102_perm_0, x = var_1145_cast_fp16_3)[name = string("transpose_4070")]; tensor reshape_153_cast_fp16 = reshape(shape = concat_514, x = transpose_102_cast_fp16)[name = string("reshape_153_cast_fp16")]; tensor transpose_103_perm_0 = const()[name = string("transpose_103_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_515 = const()[name = string("concat_515"), val = tensor([1, 64, 104])]; tensor transpose_103_cast_fp16 = transpose(perm = transpose_103_perm_0, x = var_1162_cast_fp16_3)[name = string("transpose_4069")]; tensor reshape_154_cast_fp16 = reshape(shape = concat_515, x = transpose_103_cast_fp16)[name = string("reshape_154_cast_fp16")]; bool matmul_51_transpose_x_0 = const()[name = string("matmul_51_transpose_x_0"), val = bool(false)]; bool matmul_51_transpose_y_0 = const()[name = string("matmul_51_transpose_y_0"), val = bool(false)]; tensor matmul_51_cast_fp16 = matmul(transpose_x = matmul_51_transpose_x_0, transpose_y = matmul_51_transpose_y_0, x = reshape_153_cast_fp16, y = reshape_154_cast_fp16)[name = string("matmul_51_cast_fp16")]; tensor concat_519 = const()[name = string("concat_519"), val = tensor([1, 1, 104, 104])]; tensor reshape_155_cast_fp16 = reshape(shape = concat_519, x = matmul_51_cast_fp16)[name = string("reshape_155_cast_fp16")]; tensor transpose_2739_perm_0 = const()[name = string("transpose_2739_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2739 = transpose(perm = transpose_2739_perm_0, x = reshape_155_cast_fp16)[name = string("transpose_4068")]; tensor w_207_cast_fp16 = add(x = transpose_2739, y = transpose_2305)[name = string("w_207_cast_fp16")]; tensor var_1225_cast_fp16 = softmax(axis = var_1089, x = w_207_cast_fp16)[name = string("op_1225_cast_fp16")]; string var_1227_equation_0 = const()[name = string("op_1227_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1227_cast_fp16 = einsum(equation = var_1227_equation_0, values = (var_1179_cast_fp16_3, var_1225_cast_fp16))[name = string("op_1227_cast_fp16")]; tensor transpose_104_perm_0 = const()[name = string("transpose_104_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_524 = const()[name = string("concat_524"), val = tensor([1, 104, 64])]; tensor transpose_104_cast_fp16 = transpose(perm = transpose_104_perm_0, x = var_1145_cast_fp16_4)[name = string("transpose_4067")]; tensor reshape_156_cast_fp16 = reshape(shape = concat_524, x = transpose_104_cast_fp16)[name = string("reshape_156_cast_fp16")]; tensor transpose_105_perm_0 = const()[name = string("transpose_105_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_525 = const()[name = string("concat_525"), val = tensor([1, 64, 104])]; tensor transpose_105_cast_fp16 = transpose(perm = transpose_105_perm_0, x = var_1162_cast_fp16_4)[name = string("transpose_4066")]; tensor reshape_157_cast_fp16 = reshape(shape = concat_525, x = transpose_105_cast_fp16)[name = string("reshape_157_cast_fp16")]; bool matmul_52_transpose_x_0 = const()[name = string("matmul_52_transpose_x_0"), val = bool(false)]; bool matmul_52_transpose_y_0 = const()[name = string("matmul_52_transpose_y_0"), val = bool(false)]; tensor matmul_52_cast_fp16 = matmul(transpose_x = matmul_52_transpose_x_0, transpose_y = matmul_52_transpose_y_0, x = reshape_156_cast_fp16, y = reshape_157_cast_fp16)[name = string("matmul_52_cast_fp16")]; tensor concat_529 = const()[name = string("concat_529"), val = tensor([1, 1, 104, 104])]; tensor reshape_158_cast_fp16 = reshape(shape = concat_529, x = matmul_52_cast_fp16)[name = string("reshape_158_cast_fp16")]; tensor transpose_2740_perm_0 = const()[name = string("transpose_2740_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2740 = transpose(perm = transpose_2740_perm_0, x = reshape_158_cast_fp16)[name = string("transpose_4065")]; tensor w_211_cast_fp16 = add(x = transpose_2740, y = transpose_2305)[name = string("w_211_cast_fp16")]; tensor var_1233_cast_fp16 = softmax(axis = var_1089, x = w_211_cast_fp16)[name = string("op_1233_cast_fp16")]; string var_1235_equation_0 = const()[name = string("op_1235_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1235_cast_fp16 = einsum(equation = var_1235_equation_0, values = (var_1179_cast_fp16_4, var_1233_cast_fp16))[name = string("op_1235_cast_fp16")]; tensor transpose_106_perm_0 = const()[name = string("transpose_106_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_534 = const()[name = string("concat_534"), val = tensor([1, 104, 64])]; tensor transpose_106_cast_fp16 = transpose(perm = transpose_106_perm_0, x = var_1145_cast_fp16_5)[name = string("transpose_4064")]; tensor reshape_159_cast_fp16 = reshape(shape = concat_534, x = transpose_106_cast_fp16)[name = string("reshape_159_cast_fp16")]; tensor transpose_107_perm_0 = const()[name = string("transpose_107_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_535 = const()[name = string("concat_535"), val = tensor([1, 64, 104])]; tensor transpose_107_cast_fp16 = transpose(perm = transpose_107_perm_0, x = var_1162_cast_fp16_5)[name = string("transpose_4063")]; tensor reshape_160_cast_fp16 = reshape(shape = concat_535, x = transpose_107_cast_fp16)[name = string("reshape_160_cast_fp16")]; bool matmul_53_transpose_x_0 = const()[name = string("matmul_53_transpose_x_0"), val = bool(false)]; bool matmul_53_transpose_y_0 = const()[name = string("matmul_53_transpose_y_0"), val = bool(false)]; tensor matmul_53_cast_fp16 = matmul(transpose_x = matmul_53_transpose_x_0, transpose_y = matmul_53_transpose_y_0, x = reshape_159_cast_fp16, y = reshape_160_cast_fp16)[name = string("matmul_53_cast_fp16")]; tensor concat_539 = const()[name = string("concat_539"), val = tensor([1, 1, 104, 104])]; tensor reshape_161_cast_fp16 = reshape(shape = concat_539, x = matmul_53_cast_fp16)[name = string("reshape_161_cast_fp16")]; tensor transpose_2741_perm_0 = const()[name = string("transpose_2741_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2741 = transpose(perm = transpose_2741_perm_0, x = reshape_161_cast_fp16)[name = string("transpose_4062")]; tensor w_215_cast_fp16 = add(x = transpose_2741, y = transpose_2305)[name = string("w_215_cast_fp16")]; tensor var_1241_cast_fp16 = softmax(axis = var_1089, x = w_215_cast_fp16)[name = string("op_1241_cast_fp16")]; string var_1243_equation_0 = const()[name = string("op_1243_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1243_cast_fp16 = einsum(equation = var_1243_equation_0, values = (var_1179_cast_fp16_5, var_1241_cast_fp16))[name = string("op_1243_cast_fp16")]; tensor transpose_108_perm_0 = const()[name = string("transpose_108_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_544 = const()[name = string("concat_544"), val = tensor([1, 104, 64])]; tensor transpose_108_cast_fp16 = transpose(perm = transpose_108_perm_0, x = var_1145_cast_fp16_6)[name = string("transpose_4061")]; tensor reshape_162_cast_fp16 = reshape(shape = concat_544, x = transpose_108_cast_fp16)[name = string("reshape_162_cast_fp16")]; tensor transpose_109_perm_0 = const()[name = string("transpose_109_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_545 = const()[name = string("concat_545"), val = tensor([1, 64, 104])]; tensor transpose_109_cast_fp16 = transpose(perm = transpose_109_perm_0, x = var_1162_cast_fp16_6)[name = string("transpose_4060")]; tensor reshape_163_cast_fp16 = reshape(shape = concat_545, x = transpose_109_cast_fp16)[name = string("reshape_163_cast_fp16")]; bool matmul_54_transpose_x_0 = const()[name = string("matmul_54_transpose_x_0"), val = bool(false)]; bool matmul_54_transpose_y_0 = const()[name = string("matmul_54_transpose_y_0"), val = bool(false)]; tensor matmul_54_cast_fp16 = matmul(transpose_x = matmul_54_transpose_x_0, transpose_y = matmul_54_transpose_y_0, x = reshape_162_cast_fp16, y = reshape_163_cast_fp16)[name = string("matmul_54_cast_fp16")]; tensor concat_549 = const()[name = string("concat_549"), val = tensor([1, 1, 104, 104])]; tensor reshape_164_cast_fp16 = reshape(shape = concat_549, x = matmul_54_cast_fp16)[name = string("reshape_164_cast_fp16")]; tensor transpose_2742_perm_0 = const()[name = string("transpose_2742_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2742 = transpose(perm = transpose_2742_perm_0, x = reshape_164_cast_fp16)[name = string("transpose_4059")]; tensor w_219_cast_fp16 = add(x = transpose_2742, y = transpose_2305)[name = string("w_219_cast_fp16")]; tensor var_1249_cast_fp16 = softmax(axis = var_1089, x = w_219_cast_fp16)[name = string("op_1249_cast_fp16")]; string var_1251_equation_0 = const()[name = string("op_1251_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1251_cast_fp16 = einsum(equation = var_1251_equation_0, values = (var_1179_cast_fp16_6, var_1249_cast_fp16))[name = string("op_1251_cast_fp16")]; tensor transpose_110_perm_0 = const()[name = string("transpose_110_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_554 = const()[name = string("concat_554"), val = tensor([1, 104, 64])]; tensor transpose_110_cast_fp16 = transpose(perm = transpose_110_perm_0, x = var_1145_cast_fp16_7)[name = string("transpose_4058")]; tensor reshape_165_cast_fp16 = reshape(shape = concat_554, x = transpose_110_cast_fp16)[name = string("reshape_165_cast_fp16")]; tensor transpose_111_perm_0 = const()[name = string("transpose_111_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_555 = const()[name = string("concat_555"), val = tensor([1, 64, 104])]; tensor transpose_111_cast_fp16 = transpose(perm = transpose_111_perm_0, x = var_1162_cast_fp16_7)[name = string("transpose_4057")]; tensor reshape_166_cast_fp16 = reshape(shape = concat_555, x = transpose_111_cast_fp16)[name = string("reshape_166_cast_fp16")]; bool matmul_55_transpose_x_0 = const()[name = string("matmul_55_transpose_x_0"), val = bool(false)]; bool matmul_55_transpose_y_0 = const()[name = string("matmul_55_transpose_y_0"), val = bool(false)]; tensor matmul_55_cast_fp16 = matmul(transpose_x = matmul_55_transpose_x_0, transpose_y = matmul_55_transpose_y_0, x = reshape_165_cast_fp16, y = reshape_166_cast_fp16)[name = string("matmul_55_cast_fp16")]; tensor concat_559 = const()[name = string("concat_559"), val = tensor([1, 1, 104, 104])]; tensor reshape_167_cast_fp16 = reshape(shape = concat_559, x = matmul_55_cast_fp16)[name = string("reshape_167_cast_fp16")]; tensor transpose_2743_perm_0 = const()[name = string("transpose_2743_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2743 = transpose(perm = transpose_2743_perm_0, x = reshape_167_cast_fp16)[name = string("transpose_4056")]; tensor w_223_cast_fp16 = add(x = transpose_2743, y = transpose_2305)[name = string("w_223_cast_fp16")]; tensor var_1257_cast_fp16 = softmax(axis = var_1089, x = w_223_cast_fp16)[name = string("op_1257_cast_fp16")]; string var_1259_equation_0 = const()[name = string("op_1259_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1259_cast_fp16 = einsum(equation = var_1259_equation_0, values = (var_1179_cast_fp16_7, var_1257_cast_fp16))[name = string("op_1259_cast_fp16")]; tensor transpose_112_perm_0 = const()[name = string("transpose_112_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_564 = const()[name = string("concat_564"), val = tensor([1, 104, 64])]; tensor transpose_112_cast_fp16 = transpose(perm = transpose_112_perm_0, x = var_1145_cast_fp16_8)[name = string("transpose_4055")]; tensor reshape_168_cast_fp16 = reshape(shape = concat_564, x = transpose_112_cast_fp16)[name = string("reshape_168_cast_fp16")]; tensor transpose_113_perm_0 = const()[name = string("transpose_113_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_565 = const()[name = string("concat_565"), val = tensor([1, 64, 104])]; tensor transpose_113_cast_fp16 = transpose(perm = transpose_113_perm_0, x = var_1162_cast_fp16_8)[name = string("transpose_4054")]; tensor reshape_169_cast_fp16 = reshape(shape = concat_565, x = transpose_113_cast_fp16)[name = string("reshape_169_cast_fp16")]; bool matmul_56_transpose_x_0 = const()[name = string("matmul_56_transpose_x_0"), val = bool(false)]; bool matmul_56_transpose_y_0 = const()[name = string("matmul_56_transpose_y_0"), val = bool(false)]; tensor matmul_56_cast_fp16 = matmul(transpose_x = matmul_56_transpose_x_0, transpose_y = matmul_56_transpose_y_0, x = reshape_168_cast_fp16, y = reshape_169_cast_fp16)[name = string("matmul_56_cast_fp16")]; tensor concat_569 = const()[name = string("concat_569"), val = tensor([1, 1, 104, 104])]; tensor reshape_170_cast_fp16 = reshape(shape = concat_569, x = matmul_56_cast_fp16)[name = string("reshape_170_cast_fp16")]; tensor transpose_2744_perm_0 = const()[name = string("transpose_2744_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2744 = transpose(perm = transpose_2744_perm_0, x = reshape_170_cast_fp16)[name = string("transpose_4053")]; tensor w_227_cast_fp16 = add(x = transpose_2744, y = transpose_2305)[name = string("w_227_cast_fp16")]; tensor var_1265_cast_fp16 = softmax(axis = var_1089, x = w_227_cast_fp16)[name = string("op_1265_cast_fp16")]; string var_1267_equation_0 = const()[name = string("op_1267_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1267_cast_fp16 = einsum(equation = var_1267_equation_0, values = (var_1179_cast_fp16_8, var_1265_cast_fp16))[name = string("op_1267_cast_fp16")]; tensor transpose_114_perm_0 = const()[name = string("transpose_114_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_574 = const()[name = string("concat_574"), val = tensor([1, 104, 64])]; tensor transpose_114_cast_fp16 = transpose(perm = transpose_114_perm_0, x = var_1145_cast_fp16_9)[name = string("transpose_4052")]; tensor reshape_171_cast_fp16 = reshape(shape = concat_574, x = transpose_114_cast_fp16)[name = string("reshape_171_cast_fp16")]; tensor transpose_115_perm_0 = const()[name = string("transpose_115_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_575 = const()[name = string("concat_575"), val = tensor([1, 64, 104])]; tensor transpose_115_cast_fp16 = transpose(perm = transpose_115_perm_0, x = var_1162_cast_fp16_9)[name = string("transpose_4051")]; tensor reshape_172_cast_fp16 = reshape(shape = concat_575, x = transpose_115_cast_fp16)[name = string("reshape_172_cast_fp16")]; bool matmul_57_transpose_x_0 = const()[name = string("matmul_57_transpose_x_0"), val = bool(false)]; bool matmul_57_transpose_y_0 = const()[name = string("matmul_57_transpose_y_0"), val = bool(false)]; tensor matmul_57_cast_fp16 = matmul(transpose_x = matmul_57_transpose_x_0, transpose_y = matmul_57_transpose_y_0, x = reshape_171_cast_fp16, y = reshape_172_cast_fp16)[name = string("matmul_57_cast_fp16")]; tensor concat_579 = const()[name = string("concat_579"), val = tensor([1, 1, 104, 104])]; tensor reshape_173_cast_fp16 = reshape(shape = concat_579, x = matmul_57_cast_fp16)[name = string("reshape_173_cast_fp16")]; tensor transpose_2745_perm_0 = const()[name = string("transpose_2745_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2745 = transpose(perm = transpose_2745_perm_0, x = reshape_173_cast_fp16)[name = string("transpose_4050")]; tensor w_231_cast_fp16 = add(x = transpose_2745, y = transpose_2305)[name = string("w_231_cast_fp16")]; tensor var_1273_cast_fp16 = softmax(axis = var_1089, x = w_231_cast_fp16)[name = string("op_1273_cast_fp16")]; string var_1275_equation_0 = const()[name = string("op_1275_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1275_cast_fp16 = einsum(equation = var_1275_equation_0, values = (var_1179_cast_fp16_9, var_1273_cast_fp16))[name = string("op_1275_cast_fp16")]; tensor transpose_116_perm_0 = const()[name = string("transpose_116_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_584 = const()[name = string("concat_584"), val = tensor([1, 104, 64])]; tensor transpose_116_cast_fp16 = transpose(perm = transpose_116_perm_0, x = var_1145_cast_fp16_10)[name = string("transpose_4049")]; tensor reshape_174_cast_fp16 = reshape(shape = concat_584, x = transpose_116_cast_fp16)[name = string("reshape_174_cast_fp16")]; tensor transpose_117_perm_0 = const()[name = string("transpose_117_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_585 = const()[name = string("concat_585"), val = tensor([1, 64, 104])]; tensor transpose_117_cast_fp16 = transpose(perm = transpose_117_perm_0, x = var_1162_cast_fp16_10)[name = string("transpose_4048")]; tensor reshape_175_cast_fp16 = reshape(shape = concat_585, x = transpose_117_cast_fp16)[name = string("reshape_175_cast_fp16")]; bool matmul_58_transpose_x_0 = const()[name = string("matmul_58_transpose_x_0"), val = bool(false)]; bool matmul_58_transpose_y_0 = const()[name = string("matmul_58_transpose_y_0"), val = bool(false)]; tensor matmul_58_cast_fp16 = matmul(transpose_x = matmul_58_transpose_x_0, transpose_y = matmul_58_transpose_y_0, x = reshape_174_cast_fp16, y = reshape_175_cast_fp16)[name = string("matmul_58_cast_fp16")]; tensor concat_589 = const()[name = string("concat_589"), val = tensor([1, 1, 104, 104])]; tensor reshape_176_cast_fp16 = reshape(shape = concat_589, x = matmul_58_cast_fp16)[name = string("reshape_176_cast_fp16")]; tensor transpose_2746_perm_0 = const()[name = string("transpose_2746_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2746 = transpose(perm = transpose_2746_perm_0, x = reshape_176_cast_fp16)[name = string("transpose_4047")]; tensor w_235_cast_fp16 = add(x = transpose_2746, y = transpose_2305)[name = string("w_235_cast_fp16")]; tensor var_1281_cast_fp16 = softmax(axis = var_1089, x = w_235_cast_fp16)[name = string("op_1281_cast_fp16")]; string var_1283_equation_0 = const()[name = string("op_1283_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1283_cast_fp16 = einsum(equation = var_1283_equation_0, values = (var_1179_cast_fp16_10, var_1281_cast_fp16))[name = string("op_1283_cast_fp16")]; tensor transpose_118_perm_0 = const()[name = string("transpose_118_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_594 = const()[name = string("concat_594"), val = tensor([1, 104, 64])]; tensor transpose_118_cast_fp16 = transpose(perm = transpose_118_perm_0, x = var_1145_cast_fp16_11)[name = string("transpose_4046")]; tensor reshape_177_cast_fp16 = reshape(shape = concat_594, x = transpose_118_cast_fp16)[name = string("reshape_177_cast_fp16")]; tensor transpose_119_perm_0 = const()[name = string("transpose_119_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_595 = const()[name = string("concat_595"), val = tensor([1, 64, 104])]; tensor transpose_119_cast_fp16 = transpose(perm = transpose_119_perm_0, x = var_1162_cast_fp16_11)[name = string("transpose_4045")]; tensor reshape_178_cast_fp16 = reshape(shape = concat_595, x = transpose_119_cast_fp16)[name = string("reshape_178_cast_fp16")]; bool matmul_59_transpose_x_0 = const()[name = string("matmul_59_transpose_x_0"), val = bool(false)]; bool matmul_59_transpose_y_0 = const()[name = string("matmul_59_transpose_y_0"), val = bool(false)]; tensor matmul_59_cast_fp16 = matmul(transpose_x = matmul_59_transpose_x_0, transpose_y = matmul_59_transpose_y_0, x = reshape_177_cast_fp16, y = reshape_178_cast_fp16)[name = string("matmul_59_cast_fp16")]; tensor concat_599 = const()[name = string("concat_599"), val = tensor([1, 1, 104, 104])]; tensor reshape_179_cast_fp16 = reshape(shape = concat_599, x = matmul_59_cast_fp16)[name = string("reshape_179_cast_fp16")]; tensor transpose_2747_perm_0 = const()[name = string("transpose_2747_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2747 = transpose(perm = transpose_2747_perm_0, x = reshape_179_cast_fp16)[name = string("transpose_4044")]; tensor w_239_cast_fp16 = add(x = transpose_2747, y = transpose_2305)[name = string("w_239_cast_fp16")]; tensor var_1289_cast_fp16 = softmax(axis = var_1089, x = w_239_cast_fp16)[name = string("op_1289_cast_fp16")]; string var_1291_equation_0 = const()[name = string("op_1291_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1291_cast_fp16 = einsum(equation = var_1291_equation_0, values = (var_1179_cast_fp16_11, var_1289_cast_fp16))[name = string("op_1291_cast_fp16")]; tensor transpose_120_perm_0 = const()[name = string("transpose_120_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_604 = const()[name = string("concat_604"), val = tensor([1, 104, 64])]; tensor transpose_120_cast_fp16 = transpose(perm = transpose_120_perm_0, x = var_1145_cast_fp16_12)[name = string("transpose_4043")]; tensor reshape_180_cast_fp16 = reshape(shape = concat_604, x = transpose_120_cast_fp16)[name = string("reshape_180_cast_fp16")]; tensor transpose_121_perm_0 = const()[name = string("transpose_121_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_605 = const()[name = string("concat_605"), val = tensor([1, 64, 104])]; tensor transpose_121_cast_fp16 = transpose(perm = transpose_121_perm_0, x = var_1162_cast_fp16_12)[name = string("transpose_4042")]; tensor reshape_181_cast_fp16 = reshape(shape = concat_605, x = transpose_121_cast_fp16)[name = string("reshape_181_cast_fp16")]; bool matmul_60_transpose_x_0 = const()[name = string("matmul_60_transpose_x_0"), val = bool(false)]; bool matmul_60_transpose_y_0 = const()[name = string("matmul_60_transpose_y_0"), val = bool(false)]; tensor matmul_60_cast_fp16 = matmul(transpose_x = matmul_60_transpose_x_0, transpose_y = matmul_60_transpose_y_0, x = reshape_180_cast_fp16, y = reshape_181_cast_fp16)[name = string("matmul_60_cast_fp16")]; tensor concat_609 = const()[name = string("concat_609"), val = tensor([1, 1, 104, 104])]; tensor reshape_182_cast_fp16 = reshape(shape = concat_609, x = matmul_60_cast_fp16)[name = string("reshape_182_cast_fp16")]; tensor transpose_2748_perm_0 = const()[name = string("transpose_2748_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2748 = transpose(perm = transpose_2748_perm_0, x = reshape_182_cast_fp16)[name = string("transpose_4041")]; tensor w_243_cast_fp16 = add(x = transpose_2748, y = transpose_2305)[name = string("w_243_cast_fp16")]; tensor var_1297_cast_fp16 = softmax(axis = var_1089, x = w_243_cast_fp16)[name = string("op_1297_cast_fp16")]; string var_1299_equation_0 = const()[name = string("op_1299_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1299_cast_fp16 = einsum(equation = var_1299_equation_0, values = (var_1179_cast_fp16_12, var_1297_cast_fp16))[name = string("op_1299_cast_fp16")]; tensor transpose_122_perm_0 = const()[name = string("transpose_122_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_614 = const()[name = string("concat_614"), val = tensor([1, 104, 64])]; tensor transpose_122_cast_fp16 = transpose(perm = transpose_122_perm_0, x = var_1145_cast_fp16_13)[name = string("transpose_4040")]; tensor reshape_183_cast_fp16 = reshape(shape = concat_614, x = transpose_122_cast_fp16)[name = string("reshape_183_cast_fp16")]; tensor transpose_123_perm_0 = const()[name = string("transpose_123_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_615 = const()[name = string("concat_615"), val = tensor([1, 64, 104])]; tensor transpose_123_cast_fp16 = transpose(perm = transpose_123_perm_0, x = var_1162_cast_fp16_13)[name = string("transpose_4039")]; tensor reshape_184_cast_fp16 = reshape(shape = concat_615, x = transpose_123_cast_fp16)[name = string("reshape_184_cast_fp16")]; bool matmul_61_transpose_x_0 = const()[name = string("matmul_61_transpose_x_0"), val = bool(false)]; bool matmul_61_transpose_y_0 = const()[name = string("matmul_61_transpose_y_0"), val = bool(false)]; tensor matmul_61_cast_fp16 = matmul(transpose_x = matmul_61_transpose_x_0, transpose_y = matmul_61_transpose_y_0, x = reshape_183_cast_fp16, y = reshape_184_cast_fp16)[name = string("matmul_61_cast_fp16")]; tensor concat_619 = const()[name = string("concat_619"), val = tensor([1, 1, 104, 104])]; tensor reshape_185_cast_fp16 = reshape(shape = concat_619, x = matmul_61_cast_fp16)[name = string("reshape_185_cast_fp16")]; tensor transpose_2749_perm_0 = const()[name = string("transpose_2749_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2749 = transpose(perm = transpose_2749_perm_0, x = reshape_185_cast_fp16)[name = string("transpose_4038")]; tensor w_247_cast_fp16 = add(x = transpose_2749, y = transpose_2305)[name = string("w_247_cast_fp16")]; tensor var_1305_cast_fp16 = softmax(axis = var_1089, x = w_247_cast_fp16)[name = string("op_1305_cast_fp16")]; string var_1307_equation_0 = const()[name = string("op_1307_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1307_cast_fp16 = einsum(equation = var_1307_equation_0, values = (var_1179_cast_fp16_13, var_1305_cast_fp16))[name = string("op_1307_cast_fp16")]; tensor transpose_124_perm_0 = const()[name = string("transpose_124_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_624 = const()[name = string("concat_624"), val = tensor([1, 104, 64])]; tensor transpose_124_cast_fp16 = transpose(perm = transpose_124_perm_0, x = var_1145_cast_fp16_14)[name = string("transpose_4037")]; tensor reshape_186_cast_fp16 = reshape(shape = concat_624, x = transpose_124_cast_fp16)[name = string("reshape_186_cast_fp16")]; tensor transpose_125_perm_0 = const()[name = string("transpose_125_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_625 = const()[name = string("concat_625"), val = tensor([1, 64, 104])]; tensor transpose_125_cast_fp16 = transpose(perm = transpose_125_perm_0, x = var_1162_cast_fp16_14)[name = string("transpose_4036")]; tensor reshape_187_cast_fp16 = reshape(shape = concat_625, x = transpose_125_cast_fp16)[name = string("reshape_187_cast_fp16")]; bool matmul_62_transpose_x_0 = const()[name = string("matmul_62_transpose_x_0"), val = bool(false)]; bool matmul_62_transpose_y_0 = const()[name = string("matmul_62_transpose_y_0"), val = bool(false)]; tensor matmul_62_cast_fp16 = matmul(transpose_x = matmul_62_transpose_x_0, transpose_y = matmul_62_transpose_y_0, x = reshape_186_cast_fp16, y = reshape_187_cast_fp16)[name = string("matmul_62_cast_fp16")]; tensor concat_629 = const()[name = string("concat_629"), val = tensor([1, 1, 104, 104])]; tensor reshape_188_cast_fp16 = reshape(shape = concat_629, x = matmul_62_cast_fp16)[name = string("reshape_188_cast_fp16")]; tensor transpose_2750_perm_0 = const()[name = string("transpose_2750_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2750 = transpose(perm = transpose_2750_perm_0, x = reshape_188_cast_fp16)[name = string("transpose_4035")]; tensor w_251_cast_fp16 = add(x = transpose_2750, y = transpose_2305)[name = string("w_251_cast_fp16")]; tensor var_1313_cast_fp16 = softmax(axis = var_1089, x = w_251_cast_fp16)[name = string("op_1313_cast_fp16")]; string var_1315_equation_0 = const()[name = string("op_1315_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1315_cast_fp16 = einsum(equation = var_1315_equation_0, values = (var_1179_cast_fp16_14, var_1313_cast_fp16))[name = string("op_1315_cast_fp16")]; tensor transpose_126_perm_0 = const()[name = string("transpose_126_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_634 = const()[name = string("concat_634"), val = tensor([1, 104, 64])]; tensor transpose_126_cast_fp16 = transpose(perm = transpose_126_perm_0, x = var_1145_cast_fp16_15)[name = string("transpose_4034")]; tensor reshape_189_cast_fp16 = reshape(shape = concat_634, x = transpose_126_cast_fp16)[name = string("reshape_189_cast_fp16")]; tensor transpose_127_perm_0 = const()[name = string("transpose_127_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_635 = const()[name = string("concat_635"), val = tensor([1, 64, 104])]; tensor transpose_127_cast_fp16 = transpose(perm = transpose_127_perm_0, x = var_1162_cast_fp16_15)[name = string("transpose_4033")]; tensor reshape_190_cast_fp16 = reshape(shape = concat_635, x = transpose_127_cast_fp16)[name = string("reshape_190_cast_fp16")]; bool matmul_63_transpose_x_0 = const()[name = string("matmul_63_transpose_x_0"), val = bool(false)]; bool matmul_63_transpose_y_0 = const()[name = string("matmul_63_transpose_y_0"), val = bool(false)]; tensor matmul_63_cast_fp16 = matmul(transpose_x = matmul_63_transpose_x_0, transpose_y = matmul_63_transpose_y_0, x = reshape_189_cast_fp16, y = reshape_190_cast_fp16)[name = string("matmul_63_cast_fp16")]; tensor concat_639 = const()[name = string("concat_639"), val = tensor([1, 1, 104, 104])]; tensor reshape_191_cast_fp16 = reshape(shape = concat_639, x = matmul_63_cast_fp16)[name = string("reshape_191_cast_fp16")]; tensor transpose_2751_perm_0 = const()[name = string("transpose_2751_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2751 = transpose(perm = transpose_2751_perm_0, x = reshape_191_cast_fp16)[name = string("transpose_4032")]; tensor w_255_cast_fp16 = add(x = transpose_2751, y = transpose_2305)[name = string("w_255_cast_fp16")]; tensor var_1321_cast_fp16 = softmax(axis = var_1089, x = w_255_cast_fp16)[name = string("op_1321_cast_fp16")]; string var_1323_equation_0 = const()[name = string("op_1323_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1323_cast_fp16 = einsum(equation = var_1323_equation_0, values = (var_1179_cast_fp16_15, var_1321_cast_fp16))[name = string("op_1323_cast_fp16")]; bool input_35_interleave_0 = const()[name = string("input_35_interleave_0"), val = bool(false)]; tensor input_35_cast_fp16 = concat(axis = var_1089, interleave = input_35_interleave_0, values = (var_1203_cast_fp16, var_1211_cast_fp16, var_1219_cast_fp16, var_1227_cast_fp16, var_1235_cast_fp16, var_1243_cast_fp16, var_1251_cast_fp16, var_1259_cast_fp16, var_1267_cast_fp16, var_1275_cast_fp16, var_1283_cast_fp16, var_1291_cast_fp16, var_1299_cast_fp16, var_1307_cast_fp16, var_1315_cast_fp16, var_1323_cast_fp16))[name = string("input_35_cast_fp16")]; string var_1332_pad_type_0 = const()[name = string("op_1332_pad_type_0"), val = string("valid")]; tensor var_1332_strides_0 = const()[name = string("op_1332_strides_0"), val = tensor([1, 1])]; tensor var_1332_pad_0 = const()[name = string("op_1332_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1332_dilations_0 = const()[name = string("op_1332_dilations_0"), val = tensor([1, 1])]; int32 var_1332_groups_0 = const()[name = string("op_1332_groups_0"), val = int32(1)]; tensor layers_3_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_3_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(105950144)))]; tensor layers_3_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_3_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(108047360)))]; tensor var_1332_cast_fp16 = conv(bias = layers_3_self_attn_out_proj_bias_to_fp16, dilations = var_1332_dilations_0, groups = var_1332_groups_0, pad = var_1332_pad_0, pad_type = var_1332_pad_type_0, strides = var_1332_strides_0, weight = layers_3_self_attn_out_proj_weight_to_fp16, x = input_35_cast_fp16)[name = string("op_1332_cast_fp16")]; tensor x_47_cast_fp16 = add(x = x_43_cast_fp16, y = var_1332_cast_fp16)[name = string("x_47_cast_fp16")]; tensor mu_15_axes_0 = const()[name = string("mu_15_axes_0"), val = tensor([1])]; bool mu_15_keep_dims_0 = const()[name = string("mu_15_keep_dims_0"), val = bool(true)]; tensor mu_15_cast_fp16 = reduce_mean(axes = mu_15_axes_0, keep_dims = mu_15_keep_dims_0, x = x_47_cast_fp16)[name = string("mu_15_cast_fp16")]; tensor var_1338_cast_fp16 = sub(x = x_47_cast_fp16, y = mu_15_cast_fp16)[name = string("op_1338_cast_fp16")]; fp16 var_1092_promoted_1_to_fp16 = const()[name = string("op_1092_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_1339_cast_fp16 = pow(x = var_1338_cast_fp16, y = var_1092_promoted_1_to_fp16)[name = string("op_1339_cast_fp16")]; tensor var_15_axes_0 = const()[name = string("var_15_axes_0"), val = tensor([1])]; bool var_15_keep_dims_0 = const()[name = string("var_15_keep_dims_0"), val = bool(true)]; tensor var_15_cast_fp16 = reduce_mean(axes = var_15_axes_0, keep_dims = var_15_keep_dims_0, x = var_1339_cast_fp16)[name = string("var_15_cast_fp16")]; fp16 var_1343_to_fp16 = const()[name = string("op_1343_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1344_cast_fp16 = add(x = var_15_cast_fp16, y = var_1343_to_fp16)[name = string("op_1344_cast_fp16")]; fp32 var_1345_epsilon_0 = const()[name = string("op_1345_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1345_cast_fp16 = rsqrt(epsilon = var_1345_epsilon_0, x = var_1344_cast_fp16)[name = string("op_1345_cast_fp16")]; tensor x_49_cast_fp16 = mul(x = var_1338_cast_fp16, y = var_1345_cast_fp16)[name = string("x_49_cast_fp16")]; tensor input_37_gamma_0_to_fp16 = const()[name = string("input_37_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(108049472)))]; tensor input_37_beta_0_to_fp16 = const()[name = string("input_37_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(108051584)))]; fp16 input_37_epsilon_0_to_fp16 = const()[name = string("input_37_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_37_cast_fp16 = batch_norm(beta = input_37_beta_0_to_fp16, epsilon = input_37_epsilon_0_to_fp16, gamma = input_37_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_49_cast_fp16)[name = string("input_37_cast_fp16")]; string x_51_pad_type_0 = const()[name = string("x_51_pad_type_0"), val = string("valid")]; tensor x_51_strides_0 = const()[name = string("x_51_strides_0"), val = tensor([1, 1])]; tensor x_51_pad_0 = const()[name = string("x_51_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_51_dilations_0 = const()[name = string("x_51_dilations_0"), val = tensor([1, 1])]; int32 x_51_groups_0 = const()[name = string("x_51_groups_0"), val = int32(1)]; tensor layers_3_fc1_weight_to_fp16 = const()[name = string("layers_3_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(108053696)))]; tensor layers_3_fc1_bias_to_fp16 = const()[name = string("layers_3_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(116442368)))]; tensor x_51_cast_fp16 = conv(bias = layers_3_fc1_bias_to_fp16, dilations = x_51_dilations_0, groups = x_51_groups_0, pad = x_51_pad_0, pad_type = x_51_pad_type_0, strides = x_51_strides_0, weight = layers_3_fc1_weight_to_fp16, x = input_37_cast_fp16)[name = string("x_51_cast_fp16")]; fp16 var_1360_to_fp16 = const()[name = string("op_1360_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_1361_cast_fp16 = mul(x = x_51_cast_fp16, y = var_1360_to_fp16)[name = string("op_1361_cast_fp16")]; tensor var_1362_cast_fp16 = mul(x = var_1361_cast_fp16, y = x_51_cast_fp16)[name = string("op_1362_cast_fp16")]; tensor var_1363_cast_fp16 = mul(x = var_1362_cast_fp16, y = x_51_cast_fp16)[name = string("op_1363_cast_fp16")]; tensor var_1364_cast_fp16 = add(x = x_51_cast_fp16, y = var_1363_cast_fp16)[name = string("op_1364_cast_fp16")]; fp16 var_1365_to_fp16 = const()[name = string("op_1365_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_13_cast_fp16 = mul(x = var_1364_cast_fp16, y = var_1365_to_fp16)[name = string("u_13_cast_fp16")]; fp16 var_1367_to_fp16 = const()[name = string("op_1367_to_fp16"), val = fp16(0x1p-1)]; tensor var_1368_cast_fp16 = mul(x = x_51_cast_fp16, y = var_1367_to_fp16)[name = string("op_1368_cast_fp16")]; tensor var_1369_cast_fp16 = tanh(x = u_13_cast_fp16)[name = string("op_1369_cast_fp16")]; fp16 var_1370_to_fp16 = const()[name = string("op_1370_to_fp16"), val = fp16(0x1p+0)]; tensor var_1371_cast_fp16 = add(x = var_1369_cast_fp16, y = var_1370_to_fp16)[name = string("op_1371_cast_fp16")]; tensor input_39_cast_fp16 = mul(x = var_1368_cast_fp16, y = var_1371_cast_fp16)[name = string("input_39_cast_fp16")]; string h_7_pad_type_0 = const()[name = string("h_7_pad_type_0"), val = string("valid")]; tensor h_7_strides_0 = const()[name = string("h_7_strides_0"), val = tensor([1, 1])]; tensor h_7_pad_0 = const()[name = string("h_7_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_7_dilations_0 = const()[name = string("h_7_dilations_0"), val = tensor([1, 1])]; int32 h_7_groups_0 = const()[name = string("h_7_groups_0"), val = int32(1)]; tensor layers_3_fc2_weight_to_fp16 = const()[name = string("layers_3_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(116450624)))]; tensor layers_3_fc2_bias_to_fp16 = const()[name = string("layers_3_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(124839296)))]; tensor h_7_cast_fp16 = conv(bias = layers_3_fc2_bias_to_fp16, dilations = h_7_dilations_0, groups = h_7_groups_0, pad = h_7_pad_0, pad_type = h_7_pad_type_0, strides = h_7_strides_0, weight = layers_3_fc2_weight_to_fp16, x = input_39_cast_fp16)[name = string("h_7_cast_fp16")]; tensor x_53_cast_fp16 = add(x = x_47_cast_fp16, y = h_7_cast_fp16)[name = string("x_53_cast_fp16")]; int32 var_1387 = const()[name = string("op_1387"), val = int32(1)]; tensor mu_17_axes_0 = const()[name = string("mu_17_axes_0"), val = tensor([1])]; bool mu_17_keep_dims_0 = const()[name = string("mu_17_keep_dims_0"), val = bool(true)]; tensor mu_17_cast_fp16 = reduce_mean(axes = mu_17_axes_0, keep_dims = mu_17_keep_dims_0, x = x_53_cast_fp16)[name = string("mu_17_cast_fp16")]; tensor var_1401_cast_fp16 = sub(x = x_53_cast_fp16, y = mu_17_cast_fp16)[name = string("op_1401_cast_fp16")]; fp16 var_1390_promoted_to_fp16 = const()[name = string("op_1390_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_1402_cast_fp16 = pow(x = var_1401_cast_fp16, y = var_1390_promoted_to_fp16)[name = string("op_1402_cast_fp16")]; tensor var_17_axes_0 = const()[name = string("var_17_axes_0"), val = tensor([1])]; bool var_17_keep_dims_0 = const()[name = string("var_17_keep_dims_0"), val = bool(true)]; tensor var_17_cast_fp16 = reduce_mean(axes = var_17_axes_0, keep_dims = var_17_keep_dims_0, x = var_1402_cast_fp16)[name = string("var_17_cast_fp16")]; fp16 var_1406_to_fp16 = const()[name = string("op_1406_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1407_cast_fp16 = add(x = var_17_cast_fp16, y = var_1406_to_fp16)[name = string("op_1407_cast_fp16")]; fp32 var_1408_epsilon_0 = const()[name = string("op_1408_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1408_cast_fp16 = rsqrt(epsilon = var_1408_epsilon_0, x = var_1407_cast_fp16)[name = string("op_1408_cast_fp16")]; tensor x_55_cast_fp16 = mul(x = var_1401_cast_fp16, y = var_1408_cast_fp16)[name = string("x_55_cast_fp16")]; tensor input_41_gamma_0_to_fp16 = const()[name = string("input_41_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(124841408)))]; tensor input_41_beta_0_to_fp16 = const()[name = string("input_41_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(124843520)))]; fp16 input_41_epsilon_0_to_fp16 = const()[name = string("input_41_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_41_cast_fp16 = batch_norm(beta = input_41_beta_0_to_fp16, epsilon = input_41_epsilon_0_to_fp16, gamma = input_41_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_55_cast_fp16)[name = string("input_41_cast_fp16")]; string var_1426_pad_type_0 = const()[name = string("op_1426_pad_type_0"), val = string("valid")]; tensor var_1426_strides_0 = const()[name = string("op_1426_strides_0"), val = tensor([1, 1])]; tensor var_1426_pad_0 = const()[name = string("op_1426_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1426_dilations_0 = const()[name = string("op_1426_dilations_0"), val = tensor([1, 1])]; int32 var_1426_groups_0 = const()[name = string("op_1426_groups_0"), val = int32(1)]; tensor var_1428_weight_0_to_fp16 = const()[name = string("op_1428_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(124845632)))]; tensor var_1428_bias_0_to_fp16 = const()[name = string("op_1428_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(126942848)))]; tensor var_1428_cast_fp16 = conv(bias = var_1428_bias_0_to_fp16, dilations = var_1426_dilations_0, groups = var_1426_groups_0, pad = var_1426_pad_0, pad_type = var_1426_pad_type_0, strides = var_1426_strides_0, weight = var_1428_weight_0_to_fp16, x = input_41_cast_fp16)[name = string("op_1428_cast_fp16")]; string var_1435_pad_type_0 = const()[name = string("op_1435_pad_type_0"), val = string("valid")]; tensor var_1435_strides_0 = const()[name = string("op_1435_strides_0"), val = tensor([1, 1])]; tensor var_1435_pad_0 = const()[name = string("op_1435_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1435_dilations_0 = const()[name = string("op_1435_dilations_0"), val = tensor([1, 1])]; int32 var_1435_groups_0 = const()[name = string("op_1435_groups_0"), val = int32(1)]; tensor layers_4_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_4_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(126944960)))]; tensor layers_4_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_4_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(129042176)))]; tensor var_1435_cast_fp16 = conv(bias = layers_4_self_attn_k_proj_bias_to_fp16, dilations = var_1435_dilations_0, groups = var_1435_groups_0, pad = var_1435_pad_0, pad_type = var_1435_pad_type_0, strides = var_1435_strides_0, weight = layers_4_self_attn_k_proj_weight_to_fp16, x = input_41_cast_fp16)[name = string("op_1435_cast_fp16")]; string var_1442_pad_type_0 = const()[name = string("op_1442_pad_type_0"), val = string("valid")]; tensor var_1442_strides_0 = const()[name = string("op_1442_strides_0"), val = tensor([1, 1])]; tensor var_1442_pad_0 = const()[name = string("op_1442_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1442_dilations_0 = const()[name = string("op_1442_dilations_0"), val = tensor([1, 1])]; int32 var_1442_groups_0 = const()[name = string("op_1442_groups_0"), val = int32(1)]; tensor layers_4_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_4_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(129044288)))]; tensor layers_4_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_4_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(131141504)))]; tensor var_1442_cast_fp16 = conv(bias = layers_4_self_attn_v_proj_bias_to_fp16, dilations = var_1442_dilations_0, groups = var_1442_groups_0, pad = var_1442_pad_0, pad_type = var_1442_pad_type_0, strides = var_1442_strides_0, weight = layers_4_self_attn_v_proj_weight_to_fp16, x = input_41_cast_fp16)[name = string("op_1442_cast_fp16")]; tensor tile_12 = const()[name = string("tile_12"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(131143616)))]; int32 var_1443_axis_0 = const()[name = string("op_1443_axis_0"), val = int32(1)]; tensor var_1443_cast_fp16_0, tensor var_1443_cast_fp16_1, tensor var_1443_cast_fp16_2, tensor var_1443_cast_fp16_3, tensor var_1443_cast_fp16_4, tensor var_1443_cast_fp16_5, tensor var_1443_cast_fp16_6, tensor var_1443_cast_fp16_7, tensor var_1443_cast_fp16_8, tensor var_1443_cast_fp16_9, tensor var_1443_cast_fp16_10, tensor var_1443_cast_fp16_11, tensor var_1443_cast_fp16_12, tensor var_1443_cast_fp16_13, tensor var_1443_cast_fp16_14, tensor var_1443_cast_fp16_15 = split(axis = var_1443_axis_0, split_sizes = tile_12, x = var_1428_cast_fp16)[name = string("op_1443_cast_fp16")]; tensor tile_13 = const()[name = string("tile_13"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(131143744)))]; int32 var_1460_axis_0 = const()[name = string("op_1460_axis_0"), val = int32(1)]; tensor var_1460_cast_fp16_0, tensor var_1460_cast_fp16_1, tensor var_1460_cast_fp16_2, tensor var_1460_cast_fp16_3, tensor var_1460_cast_fp16_4, tensor var_1460_cast_fp16_5, tensor var_1460_cast_fp16_6, tensor var_1460_cast_fp16_7, tensor var_1460_cast_fp16_8, tensor var_1460_cast_fp16_9, tensor var_1460_cast_fp16_10, tensor var_1460_cast_fp16_11, tensor var_1460_cast_fp16_12, tensor var_1460_cast_fp16_13, tensor var_1460_cast_fp16_14, tensor var_1460_cast_fp16_15 = split(axis = var_1460_axis_0, split_sizes = tile_13, x = var_1435_cast_fp16)[name = string("op_1460_cast_fp16")]; tensor tile_14 = const()[name = string("tile_14"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(131143872)))]; int32 var_1477_axis_0 = const()[name = string("op_1477_axis_0"), val = int32(1)]; tensor var_1477_cast_fp16_0, tensor var_1477_cast_fp16_1, tensor var_1477_cast_fp16_2, tensor var_1477_cast_fp16_3, tensor var_1477_cast_fp16_4, tensor var_1477_cast_fp16_5, tensor var_1477_cast_fp16_6, tensor var_1477_cast_fp16_7, tensor var_1477_cast_fp16_8, tensor var_1477_cast_fp16_9, tensor var_1477_cast_fp16_10, tensor var_1477_cast_fp16_11, tensor var_1477_cast_fp16_12, tensor var_1477_cast_fp16_13, tensor var_1477_cast_fp16_14, tensor var_1477_cast_fp16_15 = split(axis = var_1477_axis_0, split_sizes = tile_14, x = var_1442_cast_fp16)[name = string("op_1477_cast_fp16")]; tensor transpose_128_perm_0 = const()[name = string("transpose_128_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_644 = const()[name = string("concat_644"), val = tensor([1, 104, 64])]; tensor transpose_128_cast_fp16 = transpose(perm = transpose_128_perm_0, x = var_1443_cast_fp16_0)[name = string("transpose_4031")]; tensor reshape_192_cast_fp16 = reshape(shape = concat_644, x = transpose_128_cast_fp16)[name = string("reshape_192_cast_fp16")]; tensor transpose_129_perm_0 = const()[name = string("transpose_129_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_645 = const()[name = string("concat_645"), val = tensor([1, 64, 104])]; tensor transpose_129_cast_fp16 = transpose(perm = transpose_129_perm_0, x = var_1460_cast_fp16_0)[name = string("transpose_4030")]; tensor reshape_193_cast_fp16 = reshape(shape = concat_645, x = transpose_129_cast_fp16)[name = string("reshape_193_cast_fp16")]; bool matmul_64_transpose_x_0 = const()[name = string("matmul_64_transpose_x_0"), val = bool(false)]; bool matmul_64_transpose_y_0 = const()[name = string("matmul_64_transpose_y_0"), val = bool(false)]; tensor matmul_64_cast_fp16 = matmul(transpose_x = matmul_64_transpose_x_0, transpose_y = matmul_64_transpose_y_0, x = reshape_192_cast_fp16, y = reshape_193_cast_fp16)[name = string("matmul_64_cast_fp16")]; tensor concat_649 = const()[name = string("concat_649"), val = tensor([1, 1, 104, 104])]; tensor reshape_194_cast_fp16 = reshape(shape = concat_649, x = matmul_64_cast_fp16)[name = string("reshape_194_cast_fp16")]; tensor transpose_2752_perm_0 = const()[name = string("transpose_2752_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2752 = transpose(perm = transpose_2752_perm_0, x = reshape_194_cast_fp16)[name = string("transpose_4029")]; tensor w_259_cast_fp16 = add(x = transpose_2752, y = transpose_2305)[name = string("w_259_cast_fp16")]; tensor var_1499_cast_fp16 = softmax(axis = var_1387, x = w_259_cast_fp16)[name = string("op_1499_cast_fp16")]; string var_1501_equation_0 = const()[name = string("op_1501_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1501_cast_fp16 = einsum(equation = var_1501_equation_0, values = (var_1477_cast_fp16_0, var_1499_cast_fp16))[name = string("op_1501_cast_fp16")]; tensor transpose_130_perm_0 = const()[name = string("transpose_130_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_654 = const()[name = string("concat_654"), val = tensor([1, 104, 64])]; tensor transpose_130_cast_fp16 = transpose(perm = transpose_130_perm_0, x = var_1443_cast_fp16_1)[name = string("transpose_4028")]; tensor reshape_195_cast_fp16 = reshape(shape = concat_654, x = transpose_130_cast_fp16)[name = string("reshape_195_cast_fp16")]; tensor transpose_131_perm_0 = const()[name = string("transpose_131_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_655 = const()[name = string("concat_655"), val = tensor([1, 64, 104])]; tensor transpose_131_cast_fp16 = transpose(perm = transpose_131_perm_0, x = var_1460_cast_fp16_1)[name = string("transpose_4027")]; tensor reshape_196_cast_fp16 = reshape(shape = concat_655, x = transpose_131_cast_fp16)[name = string("reshape_196_cast_fp16")]; bool matmul_65_transpose_x_0 = const()[name = string("matmul_65_transpose_x_0"), val = bool(false)]; bool matmul_65_transpose_y_0 = const()[name = string("matmul_65_transpose_y_0"), val = bool(false)]; tensor matmul_65_cast_fp16 = matmul(transpose_x = matmul_65_transpose_x_0, transpose_y = matmul_65_transpose_y_0, x = reshape_195_cast_fp16, y = reshape_196_cast_fp16)[name = string("matmul_65_cast_fp16")]; tensor concat_659 = const()[name = string("concat_659"), val = tensor([1, 1, 104, 104])]; tensor reshape_197_cast_fp16 = reshape(shape = concat_659, x = matmul_65_cast_fp16)[name = string("reshape_197_cast_fp16")]; tensor transpose_2753_perm_0 = const()[name = string("transpose_2753_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2753 = transpose(perm = transpose_2753_perm_0, x = reshape_197_cast_fp16)[name = string("transpose_4026")]; tensor w_263_cast_fp16 = add(x = transpose_2753, y = transpose_2305)[name = string("w_263_cast_fp16")]; tensor var_1507_cast_fp16 = softmax(axis = var_1387, x = w_263_cast_fp16)[name = string("op_1507_cast_fp16")]; string var_1509_equation_0 = const()[name = string("op_1509_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1509_cast_fp16 = einsum(equation = var_1509_equation_0, values = (var_1477_cast_fp16_1, var_1507_cast_fp16))[name = string("op_1509_cast_fp16")]; tensor transpose_132_perm_0 = const()[name = string("transpose_132_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_664 = const()[name = string("concat_664"), val = tensor([1, 104, 64])]; tensor transpose_132_cast_fp16 = transpose(perm = transpose_132_perm_0, x = var_1443_cast_fp16_2)[name = string("transpose_4025")]; tensor reshape_198_cast_fp16 = reshape(shape = concat_664, x = transpose_132_cast_fp16)[name = string("reshape_198_cast_fp16")]; tensor transpose_133_perm_0 = const()[name = string("transpose_133_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_665 = const()[name = string("concat_665"), val = tensor([1, 64, 104])]; tensor transpose_133_cast_fp16 = transpose(perm = transpose_133_perm_0, x = var_1460_cast_fp16_2)[name = string("transpose_4024")]; tensor reshape_199_cast_fp16 = reshape(shape = concat_665, x = transpose_133_cast_fp16)[name = string("reshape_199_cast_fp16")]; bool matmul_66_transpose_x_0 = const()[name = string("matmul_66_transpose_x_0"), val = bool(false)]; bool matmul_66_transpose_y_0 = const()[name = string("matmul_66_transpose_y_0"), val = bool(false)]; tensor matmul_66_cast_fp16 = matmul(transpose_x = matmul_66_transpose_x_0, transpose_y = matmul_66_transpose_y_0, x = reshape_198_cast_fp16, y = reshape_199_cast_fp16)[name = string("matmul_66_cast_fp16")]; tensor concat_669 = const()[name = string("concat_669"), val = tensor([1, 1, 104, 104])]; tensor reshape_200_cast_fp16 = reshape(shape = concat_669, x = matmul_66_cast_fp16)[name = string("reshape_200_cast_fp16")]; tensor transpose_2754_perm_0 = const()[name = string("transpose_2754_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2754 = transpose(perm = transpose_2754_perm_0, x = reshape_200_cast_fp16)[name = string("transpose_4023")]; tensor w_267_cast_fp16 = add(x = transpose_2754, y = transpose_2305)[name = string("w_267_cast_fp16")]; tensor var_1515_cast_fp16 = softmax(axis = var_1387, x = w_267_cast_fp16)[name = string("op_1515_cast_fp16")]; string var_1517_equation_0 = const()[name = string("op_1517_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1517_cast_fp16 = einsum(equation = var_1517_equation_0, values = (var_1477_cast_fp16_2, var_1515_cast_fp16))[name = string("op_1517_cast_fp16")]; tensor transpose_134_perm_0 = const()[name = string("transpose_134_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_674 = const()[name = string("concat_674"), val = tensor([1, 104, 64])]; tensor transpose_134_cast_fp16 = transpose(perm = transpose_134_perm_0, x = var_1443_cast_fp16_3)[name = string("transpose_4022")]; tensor reshape_201_cast_fp16 = reshape(shape = concat_674, x = transpose_134_cast_fp16)[name = string("reshape_201_cast_fp16")]; tensor transpose_135_perm_0 = const()[name = string("transpose_135_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_675 = const()[name = string("concat_675"), val = tensor([1, 64, 104])]; tensor transpose_135_cast_fp16 = transpose(perm = transpose_135_perm_0, x = var_1460_cast_fp16_3)[name = string("transpose_4021")]; tensor reshape_202_cast_fp16 = reshape(shape = concat_675, x = transpose_135_cast_fp16)[name = string("reshape_202_cast_fp16")]; bool matmul_67_transpose_x_0 = const()[name = string("matmul_67_transpose_x_0"), val = bool(false)]; bool matmul_67_transpose_y_0 = const()[name = string("matmul_67_transpose_y_0"), val = bool(false)]; tensor matmul_67_cast_fp16 = matmul(transpose_x = matmul_67_transpose_x_0, transpose_y = matmul_67_transpose_y_0, x = reshape_201_cast_fp16, y = reshape_202_cast_fp16)[name = string("matmul_67_cast_fp16")]; tensor concat_679 = const()[name = string("concat_679"), val = tensor([1, 1, 104, 104])]; tensor reshape_203_cast_fp16 = reshape(shape = concat_679, x = matmul_67_cast_fp16)[name = string("reshape_203_cast_fp16")]; tensor transpose_2755_perm_0 = const()[name = string("transpose_2755_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2755 = transpose(perm = transpose_2755_perm_0, x = reshape_203_cast_fp16)[name = string("transpose_4020")]; tensor w_271_cast_fp16 = add(x = transpose_2755, y = transpose_2305)[name = string("w_271_cast_fp16")]; tensor var_1523_cast_fp16 = softmax(axis = var_1387, x = w_271_cast_fp16)[name = string("op_1523_cast_fp16")]; string var_1525_equation_0 = const()[name = string("op_1525_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1525_cast_fp16 = einsum(equation = var_1525_equation_0, values = (var_1477_cast_fp16_3, var_1523_cast_fp16))[name = string("op_1525_cast_fp16")]; tensor transpose_136_perm_0 = const()[name = string("transpose_136_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_684 = const()[name = string("concat_684"), val = tensor([1, 104, 64])]; tensor transpose_136_cast_fp16 = transpose(perm = transpose_136_perm_0, x = var_1443_cast_fp16_4)[name = string("transpose_4019")]; tensor reshape_204_cast_fp16 = reshape(shape = concat_684, x = transpose_136_cast_fp16)[name = string("reshape_204_cast_fp16")]; tensor transpose_137_perm_0 = const()[name = string("transpose_137_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_685 = const()[name = string("concat_685"), val = tensor([1, 64, 104])]; tensor transpose_137_cast_fp16 = transpose(perm = transpose_137_perm_0, x = var_1460_cast_fp16_4)[name = string("transpose_4018")]; tensor reshape_205_cast_fp16 = reshape(shape = concat_685, x = transpose_137_cast_fp16)[name = string("reshape_205_cast_fp16")]; bool matmul_68_transpose_x_0 = const()[name = string("matmul_68_transpose_x_0"), val = bool(false)]; bool matmul_68_transpose_y_0 = const()[name = string("matmul_68_transpose_y_0"), val = bool(false)]; tensor matmul_68_cast_fp16 = matmul(transpose_x = matmul_68_transpose_x_0, transpose_y = matmul_68_transpose_y_0, x = reshape_204_cast_fp16, y = reshape_205_cast_fp16)[name = string("matmul_68_cast_fp16")]; tensor concat_689 = const()[name = string("concat_689"), val = tensor([1, 1, 104, 104])]; tensor reshape_206_cast_fp16 = reshape(shape = concat_689, x = matmul_68_cast_fp16)[name = string("reshape_206_cast_fp16")]; tensor transpose_2756_perm_0 = const()[name = string("transpose_2756_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2756 = transpose(perm = transpose_2756_perm_0, x = reshape_206_cast_fp16)[name = string("transpose_4017")]; tensor w_275_cast_fp16 = add(x = transpose_2756, y = transpose_2305)[name = string("w_275_cast_fp16")]; tensor var_1531_cast_fp16 = softmax(axis = var_1387, x = w_275_cast_fp16)[name = string("op_1531_cast_fp16")]; string var_1533_equation_0 = const()[name = string("op_1533_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1533_cast_fp16 = einsum(equation = var_1533_equation_0, values = (var_1477_cast_fp16_4, var_1531_cast_fp16))[name = string("op_1533_cast_fp16")]; tensor transpose_138_perm_0 = const()[name = string("transpose_138_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_694 = const()[name = string("concat_694"), val = tensor([1, 104, 64])]; tensor transpose_138_cast_fp16 = transpose(perm = transpose_138_perm_0, x = var_1443_cast_fp16_5)[name = string("transpose_4016")]; tensor reshape_207_cast_fp16 = reshape(shape = concat_694, x = transpose_138_cast_fp16)[name = string("reshape_207_cast_fp16")]; tensor transpose_139_perm_0 = const()[name = string("transpose_139_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_695 = const()[name = string("concat_695"), val = tensor([1, 64, 104])]; tensor transpose_139_cast_fp16 = transpose(perm = transpose_139_perm_0, x = var_1460_cast_fp16_5)[name = string("transpose_4015")]; tensor reshape_208_cast_fp16 = reshape(shape = concat_695, x = transpose_139_cast_fp16)[name = string("reshape_208_cast_fp16")]; bool matmul_69_transpose_x_0 = const()[name = string("matmul_69_transpose_x_0"), val = bool(false)]; bool matmul_69_transpose_y_0 = const()[name = string("matmul_69_transpose_y_0"), val = bool(false)]; tensor matmul_69_cast_fp16 = matmul(transpose_x = matmul_69_transpose_x_0, transpose_y = matmul_69_transpose_y_0, x = reshape_207_cast_fp16, y = reshape_208_cast_fp16)[name = string("matmul_69_cast_fp16")]; tensor concat_699 = const()[name = string("concat_699"), val = tensor([1, 1, 104, 104])]; tensor reshape_209_cast_fp16 = reshape(shape = concat_699, x = matmul_69_cast_fp16)[name = string("reshape_209_cast_fp16")]; tensor transpose_2757_perm_0 = const()[name = string("transpose_2757_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2757 = transpose(perm = transpose_2757_perm_0, x = reshape_209_cast_fp16)[name = string("transpose_4014")]; tensor w_279_cast_fp16 = add(x = transpose_2757, y = transpose_2305)[name = string("w_279_cast_fp16")]; tensor var_1539_cast_fp16 = softmax(axis = var_1387, x = w_279_cast_fp16)[name = string("op_1539_cast_fp16")]; string var_1541_equation_0 = const()[name = string("op_1541_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1541_cast_fp16 = einsum(equation = var_1541_equation_0, values = (var_1477_cast_fp16_5, var_1539_cast_fp16))[name = string("op_1541_cast_fp16")]; tensor transpose_140_perm_0 = const()[name = string("transpose_140_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_704 = const()[name = string("concat_704"), val = tensor([1, 104, 64])]; tensor transpose_140_cast_fp16 = transpose(perm = transpose_140_perm_0, x = var_1443_cast_fp16_6)[name = string("transpose_4013")]; tensor reshape_210_cast_fp16 = reshape(shape = concat_704, x = transpose_140_cast_fp16)[name = string("reshape_210_cast_fp16")]; tensor transpose_141_perm_0 = const()[name = string("transpose_141_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_705 = const()[name = string("concat_705"), val = tensor([1, 64, 104])]; tensor transpose_141_cast_fp16 = transpose(perm = transpose_141_perm_0, x = var_1460_cast_fp16_6)[name = string("transpose_4012")]; tensor reshape_211_cast_fp16 = reshape(shape = concat_705, x = transpose_141_cast_fp16)[name = string("reshape_211_cast_fp16")]; bool matmul_70_transpose_x_0 = const()[name = string("matmul_70_transpose_x_0"), val = bool(false)]; bool matmul_70_transpose_y_0 = const()[name = string("matmul_70_transpose_y_0"), val = bool(false)]; tensor matmul_70_cast_fp16 = matmul(transpose_x = matmul_70_transpose_x_0, transpose_y = matmul_70_transpose_y_0, x = reshape_210_cast_fp16, y = reshape_211_cast_fp16)[name = string("matmul_70_cast_fp16")]; tensor concat_709 = const()[name = string("concat_709"), val = tensor([1, 1, 104, 104])]; tensor reshape_212_cast_fp16 = reshape(shape = concat_709, x = matmul_70_cast_fp16)[name = string("reshape_212_cast_fp16")]; tensor transpose_2758_perm_0 = const()[name = string("transpose_2758_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2758 = transpose(perm = transpose_2758_perm_0, x = reshape_212_cast_fp16)[name = string("transpose_4011")]; tensor w_283_cast_fp16 = add(x = transpose_2758, y = transpose_2305)[name = string("w_283_cast_fp16")]; tensor var_1547_cast_fp16 = softmax(axis = var_1387, x = w_283_cast_fp16)[name = string("op_1547_cast_fp16")]; string var_1549_equation_0 = const()[name = string("op_1549_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1549_cast_fp16 = einsum(equation = var_1549_equation_0, values = (var_1477_cast_fp16_6, var_1547_cast_fp16))[name = string("op_1549_cast_fp16")]; tensor transpose_142_perm_0 = const()[name = string("transpose_142_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_714 = const()[name = string("concat_714"), val = tensor([1, 104, 64])]; tensor transpose_142_cast_fp16 = transpose(perm = transpose_142_perm_0, x = var_1443_cast_fp16_7)[name = string("transpose_4010")]; tensor reshape_213_cast_fp16 = reshape(shape = concat_714, x = transpose_142_cast_fp16)[name = string("reshape_213_cast_fp16")]; tensor transpose_143_perm_0 = const()[name = string("transpose_143_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_715 = const()[name = string("concat_715"), val = tensor([1, 64, 104])]; tensor transpose_143_cast_fp16 = transpose(perm = transpose_143_perm_0, x = var_1460_cast_fp16_7)[name = string("transpose_4009")]; tensor reshape_214_cast_fp16 = reshape(shape = concat_715, x = transpose_143_cast_fp16)[name = string("reshape_214_cast_fp16")]; bool matmul_71_transpose_x_0 = const()[name = string("matmul_71_transpose_x_0"), val = bool(false)]; bool matmul_71_transpose_y_0 = const()[name = string("matmul_71_transpose_y_0"), val = bool(false)]; tensor matmul_71_cast_fp16 = matmul(transpose_x = matmul_71_transpose_x_0, transpose_y = matmul_71_transpose_y_0, x = reshape_213_cast_fp16, y = reshape_214_cast_fp16)[name = string("matmul_71_cast_fp16")]; tensor concat_719 = const()[name = string("concat_719"), val = tensor([1, 1, 104, 104])]; tensor reshape_215_cast_fp16 = reshape(shape = concat_719, x = matmul_71_cast_fp16)[name = string("reshape_215_cast_fp16")]; tensor transpose_2759_perm_0 = const()[name = string("transpose_2759_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2759 = transpose(perm = transpose_2759_perm_0, x = reshape_215_cast_fp16)[name = string("transpose_4008")]; tensor w_287_cast_fp16 = add(x = transpose_2759, y = transpose_2305)[name = string("w_287_cast_fp16")]; tensor var_1555_cast_fp16 = softmax(axis = var_1387, x = w_287_cast_fp16)[name = string("op_1555_cast_fp16")]; string var_1557_equation_0 = const()[name = string("op_1557_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1557_cast_fp16 = einsum(equation = var_1557_equation_0, values = (var_1477_cast_fp16_7, var_1555_cast_fp16))[name = string("op_1557_cast_fp16")]; tensor transpose_144_perm_0 = const()[name = string("transpose_144_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_724 = const()[name = string("concat_724"), val = tensor([1, 104, 64])]; tensor transpose_144_cast_fp16 = transpose(perm = transpose_144_perm_0, x = var_1443_cast_fp16_8)[name = string("transpose_4007")]; tensor reshape_216_cast_fp16 = reshape(shape = concat_724, x = transpose_144_cast_fp16)[name = string("reshape_216_cast_fp16")]; tensor transpose_145_perm_0 = const()[name = string("transpose_145_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_725 = const()[name = string("concat_725"), val = tensor([1, 64, 104])]; tensor transpose_145_cast_fp16 = transpose(perm = transpose_145_perm_0, x = var_1460_cast_fp16_8)[name = string("transpose_4006")]; tensor reshape_217_cast_fp16 = reshape(shape = concat_725, x = transpose_145_cast_fp16)[name = string("reshape_217_cast_fp16")]; bool matmul_72_transpose_x_0 = const()[name = string("matmul_72_transpose_x_0"), val = bool(false)]; bool matmul_72_transpose_y_0 = const()[name = string("matmul_72_transpose_y_0"), val = bool(false)]; tensor matmul_72_cast_fp16 = matmul(transpose_x = matmul_72_transpose_x_0, transpose_y = matmul_72_transpose_y_0, x = reshape_216_cast_fp16, y = reshape_217_cast_fp16)[name = string("matmul_72_cast_fp16")]; tensor concat_729 = const()[name = string("concat_729"), val = tensor([1, 1, 104, 104])]; tensor reshape_218_cast_fp16 = reshape(shape = concat_729, x = matmul_72_cast_fp16)[name = string("reshape_218_cast_fp16")]; tensor transpose_2760_perm_0 = const()[name = string("transpose_2760_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2760 = transpose(perm = transpose_2760_perm_0, x = reshape_218_cast_fp16)[name = string("transpose_4005")]; tensor w_291_cast_fp16 = add(x = transpose_2760, y = transpose_2305)[name = string("w_291_cast_fp16")]; tensor var_1563_cast_fp16 = softmax(axis = var_1387, x = w_291_cast_fp16)[name = string("op_1563_cast_fp16")]; string var_1565_equation_0 = const()[name = string("op_1565_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1565_cast_fp16 = einsum(equation = var_1565_equation_0, values = (var_1477_cast_fp16_8, var_1563_cast_fp16))[name = string("op_1565_cast_fp16")]; tensor transpose_146_perm_0 = const()[name = string("transpose_146_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_734 = const()[name = string("concat_734"), val = tensor([1, 104, 64])]; tensor transpose_146_cast_fp16 = transpose(perm = transpose_146_perm_0, x = var_1443_cast_fp16_9)[name = string("transpose_4004")]; tensor reshape_219_cast_fp16 = reshape(shape = concat_734, x = transpose_146_cast_fp16)[name = string("reshape_219_cast_fp16")]; tensor transpose_147_perm_0 = const()[name = string("transpose_147_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_735 = const()[name = string("concat_735"), val = tensor([1, 64, 104])]; tensor transpose_147_cast_fp16 = transpose(perm = transpose_147_perm_0, x = var_1460_cast_fp16_9)[name = string("transpose_4003")]; tensor reshape_220_cast_fp16 = reshape(shape = concat_735, x = transpose_147_cast_fp16)[name = string("reshape_220_cast_fp16")]; bool matmul_73_transpose_x_0 = const()[name = string("matmul_73_transpose_x_0"), val = bool(false)]; bool matmul_73_transpose_y_0 = const()[name = string("matmul_73_transpose_y_0"), val = bool(false)]; tensor matmul_73_cast_fp16 = matmul(transpose_x = matmul_73_transpose_x_0, transpose_y = matmul_73_transpose_y_0, x = reshape_219_cast_fp16, y = reshape_220_cast_fp16)[name = string("matmul_73_cast_fp16")]; tensor concat_739 = const()[name = string("concat_739"), val = tensor([1, 1, 104, 104])]; tensor reshape_221_cast_fp16 = reshape(shape = concat_739, x = matmul_73_cast_fp16)[name = string("reshape_221_cast_fp16")]; tensor transpose_2761_perm_0 = const()[name = string("transpose_2761_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2761 = transpose(perm = transpose_2761_perm_0, x = reshape_221_cast_fp16)[name = string("transpose_4002")]; tensor w_295_cast_fp16 = add(x = transpose_2761, y = transpose_2305)[name = string("w_295_cast_fp16")]; tensor var_1571_cast_fp16 = softmax(axis = var_1387, x = w_295_cast_fp16)[name = string("op_1571_cast_fp16")]; string var_1573_equation_0 = const()[name = string("op_1573_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1573_cast_fp16 = einsum(equation = var_1573_equation_0, values = (var_1477_cast_fp16_9, var_1571_cast_fp16))[name = string("op_1573_cast_fp16")]; tensor transpose_148_perm_0 = const()[name = string("transpose_148_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_744 = const()[name = string("concat_744"), val = tensor([1, 104, 64])]; tensor transpose_148_cast_fp16 = transpose(perm = transpose_148_perm_0, x = var_1443_cast_fp16_10)[name = string("transpose_4001")]; tensor reshape_222_cast_fp16 = reshape(shape = concat_744, x = transpose_148_cast_fp16)[name = string("reshape_222_cast_fp16")]; tensor transpose_149_perm_0 = const()[name = string("transpose_149_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_745 = const()[name = string("concat_745"), val = tensor([1, 64, 104])]; tensor transpose_149_cast_fp16 = transpose(perm = transpose_149_perm_0, x = var_1460_cast_fp16_10)[name = string("transpose_4000")]; tensor reshape_223_cast_fp16 = reshape(shape = concat_745, x = transpose_149_cast_fp16)[name = string("reshape_223_cast_fp16")]; bool matmul_74_transpose_x_0 = const()[name = string("matmul_74_transpose_x_0"), val = bool(false)]; bool matmul_74_transpose_y_0 = const()[name = string("matmul_74_transpose_y_0"), val = bool(false)]; tensor matmul_74_cast_fp16 = matmul(transpose_x = matmul_74_transpose_x_0, transpose_y = matmul_74_transpose_y_0, x = reshape_222_cast_fp16, y = reshape_223_cast_fp16)[name = string("matmul_74_cast_fp16")]; tensor concat_749 = const()[name = string("concat_749"), val = tensor([1, 1, 104, 104])]; tensor reshape_224_cast_fp16 = reshape(shape = concat_749, x = matmul_74_cast_fp16)[name = string("reshape_224_cast_fp16")]; tensor transpose_2762_perm_0 = const()[name = string("transpose_2762_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2762 = transpose(perm = transpose_2762_perm_0, x = reshape_224_cast_fp16)[name = string("transpose_3999")]; tensor w_299_cast_fp16 = add(x = transpose_2762, y = transpose_2305)[name = string("w_299_cast_fp16")]; tensor var_1579_cast_fp16 = softmax(axis = var_1387, x = w_299_cast_fp16)[name = string("op_1579_cast_fp16")]; string var_1581_equation_0 = const()[name = string("op_1581_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1581_cast_fp16 = einsum(equation = var_1581_equation_0, values = (var_1477_cast_fp16_10, var_1579_cast_fp16))[name = string("op_1581_cast_fp16")]; tensor transpose_150_perm_0 = const()[name = string("transpose_150_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_754 = const()[name = string("concat_754"), val = tensor([1, 104, 64])]; tensor transpose_150_cast_fp16 = transpose(perm = transpose_150_perm_0, x = var_1443_cast_fp16_11)[name = string("transpose_3998")]; tensor reshape_225_cast_fp16 = reshape(shape = concat_754, x = transpose_150_cast_fp16)[name = string("reshape_225_cast_fp16")]; tensor transpose_151_perm_0 = const()[name = string("transpose_151_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_755 = const()[name = string("concat_755"), val = tensor([1, 64, 104])]; tensor transpose_151_cast_fp16 = transpose(perm = transpose_151_perm_0, x = var_1460_cast_fp16_11)[name = string("transpose_3997")]; tensor reshape_226_cast_fp16 = reshape(shape = concat_755, x = transpose_151_cast_fp16)[name = string("reshape_226_cast_fp16")]; bool matmul_75_transpose_x_0 = const()[name = string("matmul_75_transpose_x_0"), val = bool(false)]; bool matmul_75_transpose_y_0 = const()[name = string("matmul_75_transpose_y_0"), val = bool(false)]; tensor matmul_75_cast_fp16 = matmul(transpose_x = matmul_75_transpose_x_0, transpose_y = matmul_75_transpose_y_0, x = reshape_225_cast_fp16, y = reshape_226_cast_fp16)[name = string("matmul_75_cast_fp16")]; tensor concat_759 = const()[name = string("concat_759"), val = tensor([1, 1, 104, 104])]; tensor reshape_227_cast_fp16 = reshape(shape = concat_759, x = matmul_75_cast_fp16)[name = string("reshape_227_cast_fp16")]; tensor transpose_2763_perm_0 = const()[name = string("transpose_2763_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2763 = transpose(perm = transpose_2763_perm_0, x = reshape_227_cast_fp16)[name = string("transpose_3996")]; tensor w_303_cast_fp16 = add(x = transpose_2763, y = transpose_2305)[name = string("w_303_cast_fp16")]; tensor var_1587_cast_fp16 = softmax(axis = var_1387, x = w_303_cast_fp16)[name = string("op_1587_cast_fp16")]; string var_1589_equation_0 = const()[name = string("op_1589_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1589_cast_fp16 = einsum(equation = var_1589_equation_0, values = (var_1477_cast_fp16_11, var_1587_cast_fp16))[name = string("op_1589_cast_fp16")]; tensor transpose_152_perm_0 = const()[name = string("transpose_152_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_764 = const()[name = string("concat_764"), val = tensor([1, 104, 64])]; tensor transpose_152_cast_fp16 = transpose(perm = transpose_152_perm_0, x = var_1443_cast_fp16_12)[name = string("transpose_3995")]; tensor reshape_228_cast_fp16 = reshape(shape = concat_764, x = transpose_152_cast_fp16)[name = string("reshape_228_cast_fp16")]; tensor transpose_153_perm_0 = const()[name = string("transpose_153_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_765 = const()[name = string("concat_765"), val = tensor([1, 64, 104])]; tensor transpose_153_cast_fp16 = transpose(perm = transpose_153_perm_0, x = var_1460_cast_fp16_12)[name = string("transpose_3994")]; tensor reshape_229_cast_fp16 = reshape(shape = concat_765, x = transpose_153_cast_fp16)[name = string("reshape_229_cast_fp16")]; bool matmul_76_transpose_x_0 = const()[name = string("matmul_76_transpose_x_0"), val = bool(false)]; bool matmul_76_transpose_y_0 = const()[name = string("matmul_76_transpose_y_0"), val = bool(false)]; tensor matmul_76_cast_fp16 = matmul(transpose_x = matmul_76_transpose_x_0, transpose_y = matmul_76_transpose_y_0, x = reshape_228_cast_fp16, y = reshape_229_cast_fp16)[name = string("matmul_76_cast_fp16")]; tensor concat_769 = const()[name = string("concat_769"), val = tensor([1, 1, 104, 104])]; tensor reshape_230_cast_fp16 = reshape(shape = concat_769, x = matmul_76_cast_fp16)[name = string("reshape_230_cast_fp16")]; tensor transpose_2764_perm_0 = const()[name = string("transpose_2764_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2764 = transpose(perm = transpose_2764_perm_0, x = reshape_230_cast_fp16)[name = string("transpose_3993")]; tensor w_307_cast_fp16 = add(x = transpose_2764, y = transpose_2305)[name = string("w_307_cast_fp16")]; tensor var_1595_cast_fp16 = softmax(axis = var_1387, x = w_307_cast_fp16)[name = string("op_1595_cast_fp16")]; string var_1597_equation_0 = const()[name = string("op_1597_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1597_cast_fp16 = einsum(equation = var_1597_equation_0, values = (var_1477_cast_fp16_12, var_1595_cast_fp16))[name = string("op_1597_cast_fp16")]; tensor transpose_154_perm_0 = const()[name = string("transpose_154_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_774 = const()[name = string("concat_774"), val = tensor([1, 104, 64])]; tensor transpose_154_cast_fp16 = transpose(perm = transpose_154_perm_0, x = var_1443_cast_fp16_13)[name = string("transpose_3992")]; tensor reshape_231_cast_fp16 = reshape(shape = concat_774, x = transpose_154_cast_fp16)[name = string("reshape_231_cast_fp16")]; tensor transpose_155_perm_0 = const()[name = string("transpose_155_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_775 = const()[name = string("concat_775"), val = tensor([1, 64, 104])]; tensor transpose_155_cast_fp16 = transpose(perm = transpose_155_perm_0, x = var_1460_cast_fp16_13)[name = string("transpose_3991")]; tensor reshape_232_cast_fp16 = reshape(shape = concat_775, x = transpose_155_cast_fp16)[name = string("reshape_232_cast_fp16")]; bool matmul_77_transpose_x_0 = const()[name = string("matmul_77_transpose_x_0"), val = bool(false)]; bool matmul_77_transpose_y_0 = const()[name = string("matmul_77_transpose_y_0"), val = bool(false)]; tensor matmul_77_cast_fp16 = matmul(transpose_x = matmul_77_transpose_x_0, transpose_y = matmul_77_transpose_y_0, x = reshape_231_cast_fp16, y = reshape_232_cast_fp16)[name = string("matmul_77_cast_fp16")]; tensor concat_779 = const()[name = string("concat_779"), val = tensor([1, 1, 104, 104])]; tensor reshape_233_cast_fp16 = reshape(shape = concat_779, x = matmul_77_cast_fp16)[name = string("reshape_233_cast_fp16")]; tensor transpose_2765_perm_0 = const()[name = string("transpose_2765_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2765 = transpose(perm = transpose_2765_perm_0, x = reshape_233_cast_fp16)[name = string("transpose_3990")]; tensor w_311_cast_fp16 = add(x = transpose_2765, y = transpose_2305)[name = string("w_311_cast_fp16")]; tensor var_1603_cast_fp16 = softmax(axis = var_1387, x = w_311_cast_fp16)[name = string("op_1603_cast_fp16")]; string var_1605_equation_0 = const()[name = string("op_1605_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1605_cast_fp16 = einsum(equation = var_1605_equation_0, values = (var_1477_cast_fp16_13, var_1603_cast_fp16))[name = string("op_1605_cast_fp16")]; tensor transpose_156_perm_0 = const()[name = string("transpose_156_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_784 = const()[name = string("concat_784"), val = tensor([1, 104, 64])]; tensor transpose_156_cast_fp16 = transpose(perm = transpose_156_perm_0, x = var_1443_cast_fp16_14)[name = string("transpose_3989")]; tensor reshape_234_cast_fp16 = reshape(shape = concat_784, x = transpose_156_cast_fp16)[name = string("reshape_234_cast_fp16")]; tensor transpose_157_perm_0 = const()[name = string("transpose_157_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_785 = const()[name = string("concat_785"), val = tensor([1, 64, 104])]; tensor transpose_157_cast_fp16 = transpose(perm = transpose_157_perm_0, x = var_1460_cast_fp16_14)[name = string("transpose_3988")]; tensor reshape_235_cast_fp16 = reshape(shape = concat_785, x = transpose_157_cast_fp16)[name = string("reshape_235_cast_fp16")]; bool matmul_78_transpose_x_0 = const()[name = string("matmul_78_transpose_x_0"), val = bool(false)]; bool matmul_78_transpose_y_0 = const()[name = string("matmul_78_transpose_y_0"), val = bool(false)]; tensor matmul_78_cast_fp16 = matmul(transpose_x = matmul_78_transpose_x_0, transpose_y = matmul_78_transpose_y_0, x = reshape_234_cast_fp16, y = reshape_235_cast_fp16)[name = string("matmul_78_cast_fp16")]; tensor concat_789 = const()[name = string("concat_789"), val = tensor([1, 1, 104, 104])]; tensor reshape_236_cast_fp16 = reshape(shape = concat_789, x = matmul_78_cast_fp16)[name = string("reshape_236_cast_fp16")]; tensor transpose_2766_perm_0 = const()[name = string("transpose_2766_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2766 = transpose(perm = transpose_2766_perm_0, x = reshape_236_cast_fp16)[name = string("transpose_3987")]; tensor w_315_cast_fp16 = add(x = transpose_2766, y = transpose_2305)[name = string("w_315_cast_fp16")]; tensor var_1611_cast_fp16 = softmax(axis = var_1387, x = w_315_cast_fp16)[name = string("op_1611_cast_fp16")]; string var_1613_equation_0 = const()[name = string("op_1613_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1613_cast_fp16 = einsum(equation = var_1613_equation_0, values = (var_1477_cast_fp16_14, var_1611_cast_fp16))[name = string("op_1613_cast_fp16")]; tensor transpose_158_perm_0 = const()[name = string("transpose_158_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_794 = const()[name = string("concat_794"), val = tensor([1, 104, 64])]; tensor transpose_158_cast_fp16 = transpose(perm = transpose_158_perm_0, x = var_1443_cast_fp16_15)[name = string("transpose_3986")]; tensor reshape_237_cast_fp16 = reshape(shape = concat_794, x = transpose_158_cast_fp16)[name = string("reshape_237_cast_fp16")]; tensor transpose_159_perm_0 = const()[name = string("transpose_159_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_795 = const()[name = string("concat_795"), val = tensor([1, 64, 104])]; tensor transpose_159_cast_fp16 = transpose(perm = transpose_159_perm_0, x = var_1460_cast_fp16_15)[name = string("transpose_3985")]; tensor reshape_238_cast_fp16 = reshape(shape = concat_795, x = transpose_159_cast_fp16)[name = string("reshape_238_cast_fp16")]; bool matmul_79_transpose_x_0 = const()[name = string("matmul_79_transpose_x_0"), val = bool(false)]; bool matmul_79_transpose_y_0 = const()[name = string("matmul_79_transpose_y_0"), val = bool(false)]; tensor matmul_79_cast_fp16 = matmul(transpose_x = matmul_79_transpose_x_0, transpose_y = matmul_79_transpose_y_0, x = reshape_237_cast_fp16, y = reshape_238_cast_fp16)[name = string("matmul_79_cast_fp16")]; tensor concat_799 = const()[name = string("concat_799"), val = tensor([1, 1, 104, 104])]; tensor reshape_239_cast_fp16 = reshape(shape = concat_799, x = matmul_79_cast_fp16)[name = string("reshape_239_cast_fp16")]; tensor transpose_2767_perm_0 = const()[name = string("transpose_2767_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2767 = transpose(perm = transpose_2767_perm_0, x = reshape_239_cast_fp16)[name = string("transpose_3984")]; tensor w_319_cast_fp16 = add(x = transpose_2767, y = transpose_2305)[name = string("w_319_cast_fp16")]; tensor var_1619_cast_fp16 = softmax(axis = var_1387, x = w_319_cast_fp16)[name = string("op_1619_cast_fp16")]; string var_1621_equation_0 = const()[name = string("op_1621_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1621_cast_fp16 = einsum(equation = var_1621_equation_0, values = (var_1477_cast_fp16_15, var_1619_cast_fp16))[name = string("op_1621_cast_fp16")]; bool input_43_interleave_0 = const()[name = string("input_43_interleave_0"), val = bool(false)]; tensor input_43_cast_fp16 = concat(axis = var_1387, interleave = input_43_interleave_0, values = (var_1501_cast_fp16, var_1509_cast_fp16, var_1517_cast_fp16, var_1525_cast_fp16, var_1533_cast_fp16, var_1541_cast_fp16, var_1549_cast_fp16, var_1557_cast_fp16, var_1565_cast_fp16, var_1573_cast_fp16, var_1581_cast_fp16, var_1589_cast_fp16, var_1597_cast_fp16, var_1605_cast_fp16, var_1613_cast_fp16, var_1621_cast_fp16))[name = string("input_43_cast_fp16")]; string var_1630_pad_type_0 = const()[name = string("op_1630_pad_type_0"), val = string("valid")]; tensor var_1630_strides_0 = const()[name = string("op_1630_strides_0"), val = tensor([1, 1])]; tensor var_1630_pad_0 = const()[name = string("op_1630_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1630_dilations_0 = const()[name = string("op_1630_dilations_0"), val = tensor([1, 1])]; int32 var_1630_groups_0 = const()[name = string("op_1630_groups_0"), val = int32(1)]; tensor layers_4_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_4_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(131144000)))]; tensor layers_4_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_4_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(133241216)))]; tensor var_1630_cast_fp16 = conv(bias = layers_4_self_attn_out_proj_bias_to_fp16, dilations = var_1630_dilations_0, groups = var_1630_groups_0, pad = var_1630_pad_0, pad_type = var_1630_pad_type_0, strides = var_1630_strides_0, weight = layers_4_self_attn_out_proj_weight_to_fp16, x = input_43_cast_fp16)[name = string("op_1630_cast_fp16")]; tensor x_57_cast_fp16 = add(x = x_53_cast_fp16, y = var_1630_cast_fp16)[name = string("x_57_cast_fp16")]; tensor mu_19_axes_0 = const()[name = string("mu_19_axes_0"), val = tensor([1])]; bool mu_19_keep_dims_0 = const()[name = string("mu_19_keep_dims_0"), val = bool(true)]; tensor mu_19_cast_fp16 = reduce_mean(axes = mu_19_axes_0, keep_dims = mu_19_keep_dims_0, x = x_57_cast_fp16)[name = string("mu_19_cast_fp16")]; tensor var_1636_cast_fp16 = sub(x = x_57_cast_fp16, y = mu_19_cast_fp16)[name = string("op_1636_cast_fp16")]; fp16 var_1390_promoted_1_to_fp16 = const()[name = string("op_1390_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_1637_cast_fp16 = pow(x = var_1636_cast_fp16, y = var_1390_promoted_1_to_fp16)[name = string("op_1637_cast_fp16")]; tensor var_19_axes_0 = const()[name = string("var_19_axes_0"), val = tensor([1])]; bool var_19_keep_dims_0 = const()[name = string("var_19_keep_dims_0"), val = bool(true)]; tensor var_19_cast_fp16 = reduce_mean(axes = var_19_axes_0, keep_dims = var_19_keep_dims_0, x = var_1637_cast_fp16)[name = string("var_19_cast_fp16")]; fp16 var_1641_to_fp16 = const()[name = string("op_1641_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1642_cast_fp16 = add(x = var_19_cast_fp16, y = var_1641_to_fp16)[name = string("op_1642_cast_fp16")]; fp32 var_1643_epsilon_0 = const()[name = string("op_1643_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1643_cast_fp16 = rsqrt(epsilon = var_1643_epsilon_0, x = var_1642_cast_fp16)[name = string("op_1643_cast_fp16")]; tensor x_59_cast_fp16 = mul(x = var_1636_cast_fp16, y = var_1643_cast_fp16)[name = string("x_59_cast_fp16")]; tensor input_45_gamma_0_to_fp16 = const()[name = string("input_45_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(133243328)))]; tensor input_45_beta_0_to_fp16 = const()[name = string("input_45_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(133245440)))]; fp16 input_45_epsilon_0_to_fp16 = const()[name = string("input_45_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_45_cast_fp16 = batch_norm(beta = input_45_beta_0_to_fp16, epsilon = input_45_epsilon_0_to_fp16, gamma = input_45_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_59_cast_fp16)[name = string("input_45_cast_fp16")]; string x_61_pad_type_0 = const()[name = string("x_61_pad_type_0"), val = string("valid")]; tensor x_61_strides_0 = const()[name = string("x_61_strides_0"), val = tensor([1, 1])]; tensor x_61_pad_0 = const()[name = string("x_61_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_61_dilations_0 = const()[name = string("x_61_dilations_0"), val = tensor([1, 1])]; int32 x_61_groups_0 = const()[name = string("x_61_groups_0"), val = int32(1)]; tensor layers_4_fc1_weight_to_fp16 = const()[name = string("layers_4_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(133247552)))]; tensor layers_4_fc1_bias_to_fp16 = const()[name = string("layers_4_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(141636224)))]; tensor x_61_cast_fp16 = conv(bias = layers_4_fc1_bias_to_fp16, dilations = x_61_dilations_0, groups = x_61_groups_0, pad = x_61_pad_0, pad_type = x_61_pad_type_0, strides = x_61_strides_0, weight = layers_4_fc1_weight_to_fp16, x = input_45_cast_fp16)[name = string("x_61_cast_fp16")]; fp16 var_1658_to_fp16 = const()[name = string("op_1658_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_1659_cast_fp16 = mul(x = x_61_cast_fp16, y = var_1658_to_fp16)[name = string("op_1659_cast_fp16")]; tensor var_1660_cast_fp16 = mul(x = var_1659_cast_fp16, y = x_61_cast_fp16)[name = string("op_1660_cast_fp16")]; tensor var_1661_cast_fp16 = mul(x = var_1660_cast_fp16, y = x_61_cast_fp16)[name = string("op_1661_cast_fp16")]; tensor var_1662_cast_fp16 = add(x = x_61_cast_fp16, y = var_1661_cast_fp16)[name = string("op_1662_cast_fp16")]; fp16 var_1663_to_fp16 = const()[name = string("op_1663_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_15_cast_fp16 = mul(x = var_1662_cast_fp16, y = var_1663_to_fp16)[name = string("u_15_cast_fp16")]; fp16 var_1665_to_fp16 = const()[name = string("op_1665_to_fp16"), val = fp16(0x1p-1)]; tensor var_1666_cast_fp16 = mul(x = x_61_cast_fp16, y = var_1665_to_fp16)[name = string("op_1666_cast_fp16")]; tensor var_1667_cast_fp16 = tanh(x = u_15_cast_fp16)[name = string("op_1667_cast_fp16")]; fp16 var_1668_to_fp16 = const()[name = string("op_1668_to_fp16"), val = fp16(0x1p+0)]; tensor var_1669_cast_fp16 = add(x = var_1667_cast_fp16, y = var_1668_to_fp16)[name = string("op_1669_cast_fp16")]; tensor input_47_cast_fp16 = mul(x = var_1666_cast_fp16, y = var_1669_cast_fp16)[name = string("input_47_cast_fp16")]; string h_9_pad_type_0 = const()[name = string("h_9_pad_type_0"), val = string("valid")]; tensor h_9_strides_0 = const()[name = string("h_9_strides_0"), val = tensor([1, 1])]; tensor h_9_pad_0 = const()[name = string("h_9_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_9_dilations_0 = const()[name = string("h_9_dilations_0"), val = tensor([1, 1])]; int32 h_9_groups_0 = const()[name = string("h_9_groups_0"), val = int32(1)]; tensor layers_4_fc2_weight_to_fp16 = const()[name = string("layers_4_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(141644480)))]; tensor layers_4_fc2_bias_to_fp16 = const()[name = string("layers_4_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(150033152)))]; tensor h_9_cast_fp16 = conv(bias = layers_4_fc2_bias_to_fp16, dilations = h_9_dilations_0, groups = h_9_groups_0, pad = h_9_pad_0, pad_type = h_9_pad_type_0, strides = h_9_strides_0, weight = layers_4_fc2_weight_to_fp16, x = input_47_cast_fp16)[name = string("h_9_cast_fp16")]; tensor x_63_cast_fp16 = add(x = x_57_cast_fp16, y = h_9_cast_fp16)[name = string("x_63_cast_fp16")]; int32 var_1685 = const()[name = string("op_1685"), val = int32(1)]; tensor mu_21_axes_0 = const()[name = string("mu_21_axes_0"), val = tensor([1])]; bool mu_21_keep_dims_0 = const()[name = string("mu_21_keep_dims_0"), val = bool(true)]; tensor mu_21_cast_fp16 = reduce_mean(axes = mu_21_axes_0, keep_dims = mu_21_keep_dims_0, x = x_63_cast_fp16)[name = string("mu_21_cast_fp16")]; tensor var_1699_cast_fp16 = sub(x = x_63_cast_fp16, y = mu_21_cast_fp16)[name = string("op_1699_cast_fp16")]; fp16 var_1688_promoted_to_fp16 = const()[name = string("op_1688_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_1700_cast_fp16 = pow(x = var_1699_cast_fp16, y = var_1688_promoted_to_fp16)[name = string("op_1700_cast_fp16")]; tensor var_21_axes_0 = const()[name = string("var_21_axes_0"), val = tensor([1])]; bool var_21_keep_dims_0 = const()[name = string("var_21_keep_dims_0"), val = bool(true)]; tensor var_21_cast_fp16 = reduce_mean(axes = var_21_axes_0, keep_dims = var_21_keep_dims_0, x = var_1700_cast_fp16)[name = string("var_21_cast_fp16")]; fp16 var_1704_to_fp16 = const()[name = string("op_1704_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1705_cast_fp16 = add(x = var_21_cast_fp16, y = var_1704_to_fp16)[name = string("op_1705_cast_fp16")]; fp32 var_1706_epsilon_0 = const()[name = string("op_1706_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1706_cast_fp16 = rsqrt(epsilon = var_1706_epsilon_0, x = var_1705_cast_fp16)[name = string("op_1706_cast_fp16")]; tensor x_65_cast_fp16 = mul(x = var_1699_cast_fp16, y = var_1706_cast_fp16)[name = string("x_65_cast_fp16")]; tensor input_49_gamma_0_to_fp16 = const()[name = string("input_49_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(150035264)))]; tensor input_49_beta_0_to_fp16 = const()[name = string("input_49_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(150037376)))]; fp16 input_49_epsilon_0_to_fp16 = const()[name = string("input_49_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_49_cast_fp16 = batch_norm(beta = input_49_beta_0_to_fp16, epsilon = input_49_epsilon_0_to_fp16, gamma = input_49_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_65_cast_fp16)[name = string("input_49_cast_fp16")]; string var_1724_pad_type_0 = const()[name = string("op_1724_pad_type_0"), val = string("valid")]; tensor var_1724_strides_0 = const()[name = string("op_1724_strides_0"), val = tensor([1, 1])]; tensor var_1724_pad_0 = const()[name = string("op_1724_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1724_dilations_0 = const()[name = string("op_1724_dilations_0"), val = tensor([1, 1])]; int32 var_1724_groups_0 = const()[name = string("op_1724_groups_0"), val = int32(1)]; tensor var_1726_weight_0_to_fp16 = const()[name = string("op_1726_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(150039488)))]; tensor var_1726_bias_0_to_fp16 = const()[name = string("op_1726_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(152136704)))]; tensor var_1726_cast_fp16 = conv(bias = var_1726_bias_0_to_fp16, dilations = var_1724_dilations_0, groups = var_1724_groups_0, pad = var_1724_pad_0, pad_type = var_1724_pad_type_0, strides = var_1724_strides_0, weight = var_1726_weight_0_to_fp16, x = input_49_cast_fp16)[name = string("op_1726_cast_fp16")]; string var_1733_pad_type_0 = const()[name = string("op_1733_pad_type_0"), val = string("valid")]; tensor var_1733_strides_0 = const()[name = string("op_1733_strides_0"), val = tensor([1, 1])]; tensor var_1733_pad_0 = const()[name = string("op_1733_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1733_dilations_0 = const()[name = string("op_1733_dilations_0"), val = tensor([1, 1])]; int32 var_1733_groups_0 = const()[name = string("op_1733_groups_0"), val = int32(1)]; tensor layers_5_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_5_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(152138816)))]; tensor layers_5_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_5_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(154236032)))]; tensor var_1733_cast_fp16 = conv(bias = layers_5_self_attn_k_proj_bias_to_fp16, dilations = var_1733_dilations_0, groups = var_1733_groups_0, pad = var_1733_pad_0, pad_type = var_1733_pad_type_0, strides = var_1733_strides_0, weight = layers_5_self_attn_k_proj_weight_to_fp16, x = input_49_cast_fp16)[name = string("op_1733_cast_fp16")]; string var_1740_pad_type_0 = const()[name = string("op_1740_pad_type_0"), val = string("valid")]; tensor var_1740_strides_0 = const()[name = string("op_1740_strides_0"), val = tensor([1, 1])]; tensor var_1740_pad_0 = const()[name = string("op_1740_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1740_dilations_0 = const()[name = string("op_1740_dilations_0"), val = tensor([1, 1])]; int32 var_1740_groups_0 = const()[name = string("op_1740_groups_0"), val = int32(1)]; tensor layers_5_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_5_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(154238144)))]; tensor layers_5_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_5_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(156335360)))]; tensor var_1740_cast_fp16 = conv(bias = layers_5_self_attn_v_proj_bias_to_fp16, dilations = var_1740_dilations_0, groups = var_1740_groups_0, pad = var_1740_pad_0, pad_type = var_1740_pad_type_0, strides = var_1740_strides_0, weight = layers_5_self_attn_v_proj_weight_to_fp16, x = input_49_cast_fp16)[name = string("op_1740_cast_fp16")]; tensor tile_15 = const()[name = string("tile_15"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(156337472)))]; int32 var_1741_axis_0 = const()[name = string("op_1741_axis_0"), val = int32(1)]; tensor var_1741_cast_fp16_0, tensor var_1741_cast_fp16_1, tensor var_1741_cast_fp16_2, tensor var_1741_cast_fp16_3, tensor var_1741_cast_fp16_4, tensor var_1741_cast_fp16_5, tensor var_1741_cast_fp16_6, tensor var_1741_cast_fp16_7, tensor var_1741_cast_fp16_8, tensor var_1741_cast_fp16_9, tensor var_1741_cast_fp16_10, tensor var_1741_cast_fp16_11, tensor var_1741_cast_fp16_12, tensor var_1741_cast_fp16_13, tensor var_1741_cast_fp16_14, tensor var_1741_cast_fp16_15 = split(axis = var_1741_axis_0, split_sizes = tile_15, x = var_1726_cast_fp16)[name = string("op_1741_cast_fp16")]; tensor tile_16 = const()[name = string("tile_16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(156337600)))]; int32 var_1758_axis_0 = const()[name = string("op_1758_axis_0"), val = int32(1)]; tensor var_1758_cast_fp16_0, tensor var_1758_cast_fp16_1, tensor var_1758_cast_fp16_2, tensor var_1758_cast_fp16_3, tensor var_1758_cast_fp16_4, tensor var_1758_cast_fp16_5, tensor var_1758_cast_fp16_6, tensor var_1758_cast_fp16_7, tensor var_1758_cast_fp16_8, tensor var_1758_cast_fp16_9, tensor var_1758_cast_fp16_10, tensor var_1758_cast_fp16_11, tensor var_1758_cast_fp16_12, tensor var_1758_cast_fp16_13, tensor var_1758_cast_fp16_14, tensor var_1758_cast_fp16_15 = split(axis = var_1758_axis_0, split_sizes = tile_16, x = var_1733_cast_fp16)[name = string("op_1758_cast_fp16")]; tensor tile_17 = const()[name = string("tile_17"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(156337728)))]; int32 var_1775_axis_0 = const()[name = string("op_1775_axis_0"), val = int32(1)]; tensor var_1775_cast_fp16_0, tensor var_1775_cast_fp16_1, tensor var_1775_cast_fp16_2, tensor var_1775_cast_fp16_3, tensor var_1775_cast_fp16_4, tensor var_1775_cast_fp16_5, tensor var_1775_cast_fp16_6, tensor var_1775_cast_fp16_7, tensor var_1775_cast_fp16_8, tensor var_1775_cast_fp16_9, tensor var_1775_cast_fp16_10, tensor var_1775_cast_fp16_11, tensor var_1775_cast_fp16_12, tensor var_1775_cast_fp16_13, tensor var_1775_cast_fp16_14, tensor var_1775_cast_fp16_15 = split(axis = var_1775_axis_0, split_sizes = tile_17, x = var_1740_cast_fp16)[name = string("op_1775_cast_fp16")]; tensor transpose_160_perm_0 = const()[name = string("transpose_160_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_804 = const()[name = string("concat_804"), val = tensor([1, 104, 64])]; tensor transpose_160_cast_fp16 = transpose(perm = transpose_160_perm_0, x = var_1741_cast_fp16_0)[name = string("transpose_3983")]; tensor reshape_240_cast_fp16 = reshape(shape = concat_804, x = transpose_160_cast_fp16)[name = string("reshape_240_cast_fp16")]; tensor transpose_161_perm_0 = const()[name = string("transpose_161_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_805 = const()[name = string("concat_805"), val = tensor([1, 64, 104])]; tensor transpose_161_cast_fp16 = transpose(perm = transpose_161_perm_0, x = var_1758_cast_fp16_0)[name = string("transpose_3982")]; tensor reshape_241_cast_fp16 = reshape(shape = concat_805, x = transpose_161_cast_fp16)[name = string("reshape_241_cast_fp16")]; bool matmul_80_transpose_x_0 = const()[name = string("matmul_80_transpose_x_0"), val = bool(false)]; bool matmul_80_transpose_y_0 = const()[name = string("matmul_80_transpose_y_0"), val = bool(false)]; tensor matmul_80_cast_fp16 = matmul(transpose_x = matmul_80_transpose_x_0, transpose_y = matmul_80_transpose_y_0, x = reshape_240_cast_fp16, y = reshape_241_cast_fp16)[name = string("matmul_80_cast_fp16")]; tensor concat_809 = const()[name = string("concat_809"), val = tensor([1, 1, 104, 104])]; tensor reshape_242_cast_fp16 = reshape(shape = concat_809, x = matmul_80_cast_fp16)[name = string("reshape_242_cast_fp16")]; tensor transpose_2768_perm_0 = const()[name = string("transpose_2768_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2768 = transpose(perm = transpose_2768_perm_0, x = reshape_242_cast_fp16)[name = string("transpose_3981")]; tensor w_323_cast_fp16 = add(x = transpose_2768, y = transpose_2305)[name = string("w_323_cast_fp16")]; tensor var_1797_cast_fp16 = softmax(axis = var_1685, x = w_323_cast_fp16)[name = string("op_1797_cast_fp16")]; string var_1799_equation_0 = const()[name = string("op_1799_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1799_cast_fp16 = einsum(equation = var_1799_equation_0, values = (var_1775_cast_fp16_0, var_1797_cast_fp16))[name = string("op_1799_cast_fp16")]; tensor transpose_162_perm_0 = const()[name = string("transpose_162_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_814 = const()[name = string("concat_814"), val = tensor([1, 104, 64])]; tensor transpose_162_cast_fp16 = transpose(perm = transpose_162_perm_0, x = var_1741_cast_fp16_1)[name = string("transpose_3980")]; tensor reshape_243_cast_fp16 = reshape(shape = concat_814, x = transpose_162_cast_fp16)[name = string("reshape_243_cast_fp16")]; tensor transpose_163_perm_0 = const()[name = string("transpose_163_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_815 = const()[name = string("concat_815"), val = tensor([1, 64, 104])]; tensor transpose_163_cast_fp16 = transpose(perm = transpose_163_perm_0, x = var_1758_cast_fp16_1)[name = string("transpose_3979")]; tensor reshape_244_cast_fp16 = reshape(shape = concat_815, x = transpose_163_cast_fp16)[name = string("reshape_244_cast_fp16")]; bool matmul_81_transpose_x_0 = const()[name = string("matmul_81_transpose_x_0"), val = bool(false)]; bool matmul_81_transpose_y_0 = const()[name = string("matmul_81_transpose_y_0"), val = bool(false)]; tensor matmul_81_cast_fp16 = matmul(transpose_x = matmul_81_transpose_x_0, transpose_y = matmul_81_transpose_y_0, x = reshape_243_cast_fp16, y = reshape_244_cast_fp16)[name = string("matmul_81_cast_fp16")]; tensor concat_819 = const()[name = string("concat_819"), val = tensor([1, 1, 104, 104])]; tensor reshape_245_cast_fp16 = reshape(shape = concat_819, x = matmul_81_cast_fp16)[name = string("reshape_245_cast_fp16")]; tensor transpose_2769_perm_0 = const()[name = string("transpose_2769_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2769 = transpose(perm = transpose_2769_perm_0, x = reshape_245_cast_fp16)[name = string("transpose_3978")]; tensor w_327_cast_fp16 = add(x = transpose_2769, y = transpose_2305)[name = string("w_327_cast_fp16")]; tensor var_1805_cast_fp16 = softmax(axis = var_1685, x = w_327_cast_fp16)[name = string("op_1805_cast_fp16")]; string var_1807_equation_0 = const()[name = string("op_1807_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1807_cast_fp16 = einsum(equation = var_1807_equation_0, values = (var_1775_cast_fp16_1, var_1805_cast_fp16))[name = string("op_1807_cast_fp16")]; tensor transpose_164_perm_0 = const()[name = string("transpose_164_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_824 = const()[name = string("concat_824"), val = tensor([1, 104, 64])]; tensor transpose_164_cast_fp16 = transpose(perm = transpose_164_perm_0, x = var_1741_cast_fp16_2)[name = string("transpose_3977")]; tensor reshape_246_cast_fp16 = reshape(shape = concat_824, x = transpose_164_cast_fp16)[name = string("reshape_246_cast_fp16")]; tensor transpose_165_perm_0 = const()[name = string("transpose_165_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_825 = const()[name = string("concat_825"), val = tensor([1, 64, 104])]; tensor transpose_165_cast_fp16 = transpose(perm = transpose_165_perm_0, x = var_1758_cast_fp16_2)[name = string("transpose_3976")]; tensor reshape_247_cast_fp16 = reshape(shape = concat_825, x = transpose_165_cast_fp16)[name = string("reshape_247_cast_fp16")]; bool matmul_82_transpose_x_0 = const()[name = string("matmul_82_transpose_x_0"), val = bool(false)]; bool matmul_82_transpose_y_0 = const()[name = string("matmul_82_transpose_y_0"), val = bool(false)]; tensor matmul_82_cast_fp16 = matmul(transpose_x = matmul_82_transpose_x_0, transpose_y = matmul_82_transpose_y_0, x = reshape_246_cast_fp16, y = reshape_247_cast_fp16)[name = string("matmul_82_cast_fp16")]; tensor concat_829 = const()[name = string("concat_829"), val = tensor([1, 1, 104, 104])]; tensor reshape_248_cast_fp16 = reshape(shape = concat_829, x = matmul_82_cast_fp16)[name = string("reshape_248_cast_fp16")]; tensor transpose_2770_perm_0 = const()[name = string("transpose_2770_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2770 = transpose(perm = transpose_2770_perm_0, x = reshape_248_cast_fp16)[name = string("transpose_3975")]; tensor w_331_cast_fp16 = add(x = transpose_2770, y = transpose_2305)[name = string("w_331_cast_fp16")]; tensor var_1813_cast_fp16 = softmax(axis = var_1685, x = w_331_cast_fp16)[name = string("op_1813_cast_fp16")]; string var_1815_equation_0 = const()[name = string("op_1815_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1815_cast_fp16 = einsum(equation = var_1815_equation_0, values = (var_1775_cast_fp16_2, var_1813_cast_fp16))[name = string("op_1815_cast_fp16")]; tensor transpose_166_perm_0 = const()[name = string("transpose_166_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_834 = const()[name = string("concat_834"), val = tensor([1, 104, 64])]; tensor transpose_166_cast_fp16 = transpose(perm = transpose_166_perm_0, x = var_1741_cast_fp16_3)[name = string("transpose_3974")]; tensor reshape_249_cast_fp16 = reshape(shape = concat_834, x = transpose_166_cast_fp16)[name = string("reshape_249_cast_fp16")]; tensor transpose_167_perm_0 = const()[name = string("transpose_167_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_835 = const()[name = string("concat_835"), val = tensor([1, 64, 104])]; tensor transpose_167_cast_fp16 = transpose(perm = transpose_167_perm_0, x = var_1758_cast_fp16_3)[name = string("transpose_3973")]; tensor reshape_250_cast_fp16 = reshape(shape = concat_835, x = transpose_167_cast_fp16)[name = string("reshape_250_cast_fp16")]; bool matmul_83_transpose_x_0 = const()[name = string("matmul_83_transpose_x_0"), val = bool(false)]; bool matmul_83_transpose_y_0 = const()[name = string("matmul_83_transpose_y_0"), val = bool(false)]; tensor matmul_83_cast_fp16 = matmul(transpose_x = matmul_83_transpose_x_0, transpose_y = matmul_83_transpose_y_0, x = reshape_249_cast_fp16, y = reshape_250_cast_fp16)[name = string("matmul_83_cast_fp16")]; tensor concat_839 = const()[name = string("concat_839"), val = tensor([1, 1, 104, 104])]; tensor reshape_251_cast_fp16 = reshape(shape = concat_839, x = matmul_83_cast_fp16)[name = string("reshape_251_cast_fp16")]; tensor transpose_2771_perm_0 = const()[name = string("transpose_2771_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2771 = transpose(perm = transpose_2771_perm_0, x = reshape_251_cast_fp16)[name = string("transpose_3972")]; tensor w_335_cast_fp16 = add(x = transpose_2771, y = transpose_2305)[name = string("w_335_cast_fp16")]; tensor var_1821_cast_fp16 = softmax(axis = var_1685, x = w_335_cast_fp16)[name = string("op_1821_cast_fp16")]; string var_1823_equation_0 = const()[name = string("op_1823_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1823_cast_fp16 = einsum(equation = var_1823_equation_0, values = (var_1775_cast_fp16_3, var_1821_cast_fp16))[name = string("op_1823_cast_fp16")]; tensor transpose_168_perm_0 = const()[name = string("transpose_168_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_844 = const()[name = string("concat_844"), val = tensor([1, 104, 64])]; tensor transpose_168_cast_fp16 = transpose(perm = transpose_168_perm_0, x = var_1741_cast_fp16_4)[name = string("transpose_3971")]; tensor reshape_252_cast_fp16 = reshape(shape = concat_844, x = transpose_168_cast_fp16)[name = string("reshape_252_cast_fp16")]; tensor transpose_169_perm_0 = const()[name = string("transpose_169_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_845 = const()[name = string("concat_845"), val = tensor([1, 64, 104])]; tensor transpose_169_cast_fp16 = transpose(perm = transpose_169_perm_0, x = var_1758_cast_fp16_4)[name = string("transpose_3970")]; tensor reshape_253_cast_fp16 = reshape(shape = concat_845, x = transpose_169_cast_fp16)[name = string("reshape_253_cast_fp16")]; bool matmul_84_transpose_x_0 = const()[name = string("matmul_84_transpose_x_0"), val = bool(false)]; bool matmul_84_transpose_y_0 = const()[name = string("matmul_84_transpose_y_0"), val = bool(false)]; tensor matmul_84_cast_fp16 = matmul(transpose_x = matmul_84_transpose_x_0, transpose_y = matmul_84_transpose_y_0, x = reshape_252_cast_fp16, y = reshape_253_cast_fp16)[name = string("matmul_84_cast_fp16")]; tensor concat_849 = const()[name = string("concat_849"), val = tensor([1, 1, 104, 104])]; tensor reshape_254_cast_fp16 = reshape(shape = concat_849, x = matmul_84_cast_fp16)[name = string("reshape_254_cast_fp16")]; tensor transpose_2772_perm_0 = const()[name = string("transpose_2772_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2772 = transpose(perm = transpose_2772_perm_0, x = reshape_254_cast_fp16)[name = string("transpose_3969")]; tensor w_339_cast_fp16 = add(x = transpose_2772, y = transpose_2305)[name = string("w_339_cast_fp16")]; tensor var_1829_cast_fp16 = softmax(axis = var_1685, x = w_339_cast_fp16)[name = string("op_1829_cast_fp16")]; string var_1831_equation_0 = const()[name = string("op_1831_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1831_cast_fp16 = einsum(equation = var_1831_equation_0, values = (var_1775_cast_fp16_4, var_1829_cast_fp16))[name = string("op_1831_cast_fp16")]; tensor transpose_170_perm_0 = const()[name = string("transpose_170_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_854 = const()[name = string("concat_854"), val = tensor([1, 104, 64])]; tensor transpose_170_cast_fp16 = transpose(perm = transpose_170_perm_0, x = var_1741_cast_fp16_5)[name = string("transpose_3968")]; tensor reshape_255_cast_fp16 = reshape(shape = concat_854, x = transpose_170_cast_fp16)[name = string("reshape_255_cast_fp16")]; tensor transpose_171_perm_0 = const()[name = string("transpose_171_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_855 = const()[name = string("concat_855"), val = tensor([1, 64, 104])]; tensor transpose_171_cast_fp16 = transpose(perm = transpose_171_perm_0, x = var_1758_cast_fp16_5)[name = string("transpose_3967")]; tensor reshape_256_cast_fp16 = reshape(shape = concat_855, x = transpose_171_cast_fp16)[name = string("reshape_256_cast_fp16")]; bool matmul_85_transpose_x_0 = const()[name = string("matmul_85_transpose_x_0"), val = bool(false)]; bool matmul_85_transpose_y_0 = const()[name = string("matmul_85_transpose_y_0"), val = bool(false)]; tensor matmul_85_cast_fp16 = matmul(transpose_x = matmul_85_transpose_x_0, transpose_y = matmul_85_transpose_y_0, x = reshape_255_cast_fp16, y = reshape_256_cast_fp16)[name = string("matmul_85_cast_fp16")]; tensor concat_859 = const()[name = string("concat_859"), val = tensor([1, 1, 104, 104])]; tensor reshape_257_cast_fp16 = reshape(shape = concat_859, x = matmul_85_cast_fp16)[name = string("reshape_257_cast_fp16")]; tensor transpose_2773_perm_0 = const()[name = string("transpose_2773_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2773 = transpose(perm = transpose_2773_perm_0, x = reshape_257_cast_fp16)[name = string("transpose_3966")]; tensor w_343_cast_fp16 = add(x = transpose_2773, y = transpose_2305)[name = string("w_343_cast_fp16")]; tensor var_1837_cast_fp16 = softmax(axis = var_1685, x = w_343_cast_fp16)[name = string("op_1837_cast_fp16")]; string var_1839_equation_0 = const()[name = string("op_1839_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1839_cast_fp16 = einsum(equation = var_1839_equation_0, values = (var_1775_cast_fp16_5, var_1837_cast_fp16))[name = string("op_1839_cast_fp16")]; tensor transpose_172_perm_0 = const()[name = string("transpose_172_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_864 = const()[name = string("concat_864"), val = tensor([1, 104, 64])]; tensor transpose_172_cast_fp16 = transpose(perm = transpose_172_perm_0, x = var_1741_cast_fp16_6)[name = string("transpose_3965")]; tensor reshape_258_cast_fp16 = reshape(shape = concat_864, x = transpose_172_cast_fp16)[name = string("reshape_258_cast_fp16")]; tensor transpose_173_perm_0 = const()[name = string("transpose_173_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_865 = const()[name = string("concat_865"), val = tensor([1, 64, 104])]; tensor transpose_173_cast_fp16 = transpose(perm = transpose_173_perm_0, x = var_1758_cast_fp16_6)[name = string("transpose_3964")]; tensor reshape_259_cast_fp16 = reshape(shape = concat_865, x = transpose_173_cast_fp16)[name = string("reshape_259_cast_fp16")]; bool matmul_86_transpose_x_0 = const()[name = string("matmul_86_transpose_x_0"), val = bool(false)]; bool matmul_86_transpose_y_0 = const()[name = string("matmul_86_transpose_y_0"), val = bool(false)]; tensor matmul_86_cast_fp16 = matmul(transpose_x = matmul_86_transpose_x_0, transpose_y = matmul_86_transpose_y_0, x = reshape_258_cast_fp16, y = reshape_259_cast_fp16)[name = string("matmul_86_cast_fp16")]; tensor concat_869 = const()[name = string("concat_869"), val = tensor([1, 1, 104, 104])]; tensor reshape_260_cast_fp16 = reshape(shape = concat_869, x = matmul_86_cast_fp16)[name = string("reshape_260_cast_fp16")]; tensor transpose_2774_perm_0 = const()[name = string("transpose_2774_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2774 = transpose(perm = transpose_2774_perm_0, x = reshape_260_cast_fp16)[name = string("transpose_3963")]; tensor w_347_cast_fp16 = add(x = transpose_2774, y = transpose_2305)[name = string("w_347_cast_fp16")]; tensor var_1845_cast_fp16 = softmax(axis = var_1685, x = w_347_cast_fp16)[name = string("op_1845_cast_fp16")]; string var_1847_equation_0 = const()[name = string("op_1847_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1847_cast_fp16 = einsum(equation = var_1847_equation_0, values = (var_1775_cast_fp16_6, var_1845_cast_fp16))[name = string("op_1847_cast_fp16")]; tensor transpose_174_perm_0 = const()[name = string("transpose_174_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_874 = const()[name = string("concat_874"), val = tensor([1, 104, 64])]; tensor transpose_174_cast_fp16 = transpose(perm = transpose_174_perm_0, x = var_1741_cast_fp16_7)[name = string("transpose_3962")]; tensor reshape_261_cast_fp16 = reshape(shape = concat_874, x = transpose_174_cast_fp16)[name = string("reshape_261_cast_fp16")]; tensor transpose_175_perm_0 = const()[name = string("transpose_175_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_875 = const()[name = string("concat_875"), val = tensor([1, 64, 104])]; tensor transpose_175_cast_fp16 = transpose(perm = transpose_175_perm_0, x = var_1758_cast_fp16_7)[name = string("transpose_3961")]; tensor reshape_262_cast_fp16 = reshape(shape = concat_875, x = transpose_175_cast_fp16)[name = string("reshape_262_cast_fp16")]; bool matmul_87_transpose_x_0 = const()[name = string("matmul_87_transpose_x_0"), val = bool(false)]; bool matmul_87_transpose_y_0 = const()[name = string("matmul_87_transpose_y_0"), val = bool(false)]; tensor matmul_87_cast_fp16 = matmul(transpose_x = matmul_87_transpose_x_0, transpose_y = matmul_87_transpose_y_0, x = reshape_261_cast_fp16, y = reshape_262_cast_fp16)[name = string("matmul_87_cast_fp16")]; tensor concat_879 = const()[name = string("concat_879"), val = tensor([1, 1, 104, 104])]; tensor reshape_263_cast_fp16 = reshape(shape = concat_879, x = matmul_87_cast_fp16)[name = string("reshape_263_cast_fp16")]; tensor transpose_2775_perm_0 = const()[name = string("transpose_2775_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2775 = transpose(perm = transpose_2775_perm_0, x = reshape_263_cast_fp16)[name = string("transpose_3960")]; tensor w_351_cast_fp16 = add(x = transpose_2775, y = transpose_2305)[name = string("w_351_cast_fp16")]; tensor var_1853_cast_fp16 = softmax(axis = var_1685, x = w_351_cast_fp16)[name = string("op_1853_cast_fp16")]; string var_1855_equation_0 = const()[name = string("op_1855_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1855_cast_fp16 = einsum(equation = var_1855_equation_0, values = (var_1775_cast_fp16_7, var_1853_cast_fp16))[name = string("op_1855_cast_fp16")]; tensor transpose_176_perm_0 = const()[name = string("transpose_176_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_884 = const()[name = string("concat_884"), val = tensor([1, 104, 64])]; tensor transpose_176_cast_fp16 = transpose(perm = transpose_176_perm_0, x = var_1741_cast_fp16_8)[name = string("transpose_3959")]; tensor reshape_264_cast_fp16 = reshape(shape = concat_884, x = transpose_176_cast_fp16)[name = string("reshape_264_cast_fp16")]; tensor transpose_177_perm_0 = const()[name = string("transpose_177_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_885 = const()[name = string("concat_885"), val = tensor([1, 64, 104])]; tensor transpose_177_cast_fp16 = transpose(perm = transpose_177_perm_0, x = var_1758_cast_fp16_8)[name = string("transpose_3958")]; tensor reshape_265_cast_fp16 = reshape(shape = concat_885, x = transpose_177_cast_fp16)[name = string("reshape_265_cast_fp16")]; bool matmul_88_transpose_x_0 = const()[name = string("matmul_88_transpose_x_0"), val = bool(false)]; bool matmul_88_transpose_y_0 = const()[name = string("matmul_88_transpose_y_0"), val = bool(false)]; tensor matmul_88_cast_fp16 = matmul(transpose_x = matmul_88_transpose_x_0, transpose_y = matmul_88_transpose_y_0, x = reshape_264_cast_fp16, y = reshape_265_cast_fp16)[name = string("matmul_88_cast_fp16")]; tensor concat_889 = const()[name = string("concat_889"), val = tensor([1, 1, 104, 104])]; tensor reshape_266_cast_fp16 = reshape(shape = concat_889, x = matmul_88_cast_fp16)[name = string("reshape_266_cast_fp16")]; tensor transpose_2776_perm_0 = const()[name = string("transpose_2776_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2776 = transpose(perm = transpose_2776_perm_0, x = reshape_266_cast_fp16)[name = string("transpose_3957")]; tensor w_355_cast_fp16 = add(x = transpose_2776, y = transpose_2305)[name = string("w_355_cast_fp16")]; tensor var_1861_cast_fp16 = softmax(axis = var_1685, x = w_355_cast_fp16)[name = string("op_1861_cast_fp16")]; string var_1863_equation_0 = const()[name = string("op_1863_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1863_cast_fp16 = einsum(equation = var_1863_equation_0, values = (var_1775_cast_fp16_8, var_1861_cast_fp16))[name = string("op_1863_cast_fp16")]; tensor transpose_178_perm_0 = const()[name = string("transpose_178_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_894 = const()[name = string("concat_894"), val = tensor([1, 104, 64])]; tensor transpose_178_cast_fp16 = transpose(perm = transpose_178_perm_0, x = var_1741_cast_fp16_9)[name = string("transpose_3956")]; tensor reshape_267_cast_fp16 = reshape(shape = concat_894, x = transpose_178_cast_fp16)[name = string("reshape_267_cast_fp16")]; tensor transpose_179_perm_0 = const()[name = string("transpose_179_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_895 = const()[name = string("concat_895"), val = tensor([1, 64, 104])]; tensor transpose_179_cast_fp16 = transpose(perm = transpose_179_perm_0, x = var_1758_cast_fp16_9)[name = string("transpose_3955")]; tensor reshape_268_cast_fp16 = reshape(shape = concat_895, x = transpose_179_cast_fp16)[name = string("reshape_268_cast_fp16")]; bool matmul_89_transpose_x_0 = const()[name = string("matmul_89_transpose_x_0"), val = bool(false)]; bool matmul_89_transpose_y_0 = const()[name = string("matmul_89_transpose_y_0"), val = bool(false)]; tensor matmul_89_cast_fp16 = matmul(transpose_x = matmul_89_transpose_x_0, transpose_y = matmul_89_transpose_y_0, x = reshape_267_cast_fp16, y = reshape_268_cast_fp16)[name = string("matmul_89_cast_fp16")]; tensor concat_899 = const()[name = string("concat_899"), val = tensor([1, 1, 104, 104])]; tensor reshape_269_cast_fp16 = reshape(shape = concat_899, x = matmul_89_cast_fp16)[name = string("reshape_269_cast_fp16")]; tensor transpose_2777_perm_0 = const()[name = string("transpose_2777_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2777 = transpose(perm = transpose_2777_perm_0, x = reshape_269_cast_fp16)[name = string("transpose_3954")]; tensor w_359_cast_fp16 = add(x = transpose_2777, y = transpose_2305)[name = string("w_359_cast_fp16")]; tensor var_1869_cast_fp16 = softmax(axis = var_1685, x = w_359_cast_fp16)[name = string("op_1869_cast_fp16")]; string var_1871_equation_0 = const()[name = string("op_1871_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1871_cast_fp16 = einsum(equation = var_1871_equation_0, values = (var_1775_cast_fp16_9, var_1869_cast_fp16))[name = string("op_1871_cast_fp16")]; tensor transpose_180_perm_0 = const()[name = string("transpose_180_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_904 = const()[name = string("concat_904"), val = tensor([1, 104, 64])]; tensor transpose_180_cast_fp16 = transpose(perm = transpose_180_perm_0, x = var_1741_cast_fp16_10)[name = string("transpose_3953")]; tensor reshape_270_cast_fp16 = reshape(shape = concat_904, x = transpose_180_cast_fp16)[name = string("reshape_270_cast_fp16")]; tensor transpose_181_perm_0 = const()[name = string("transpose_181_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_905 = const()[name = string("concat_905"), val = tensor([1, 64, 104])]; tensor transpose_181_cast_fp16 = transpose(perm = transpose_181_perm_0, x = var_1758_cast_fp16_10)[name = string("transpose_3952")]; tensor reshape_271_cast_fp16 = reshape(shape = concat_905, x = transpose_181_cast_fp16)[name = string("reshape_271_cast_fp16")]; bool matmul_90_transpose_x_0 = const()[name = string("matmul_90_transpose_x_0"), val = bool(false)]; bool matmul_90_transpose_y_0 = const()[name = string("matmul_90_transpose_y_0"), val = bool(false)]; tensor matmul_90_cast_fp16 = matmul(transpose_x = matmul_90_transpose_x_0, transpose_y = matmul_90_transpose_y_0, x = reshape_270_cast_fp16, y = reshape_271_cast_fp16)[name = string("matmul_90_cast_fp16")]; tensor concat_909 = const()[name = string("concat_909"), val = tensor([1, 1, 104, 104])]; tensor reshape_272_cast_fp16 = reshape(shape = concat_909, x = matmul_90_cast_fp16)[name = string("reshape_272_cast_fp16")]; tensor transpose_2778_perm_0 = const()[name = string("transpose_2778_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2778 = transpose(perm = transpose_2778_perm_0, x = reshape_272_cast_fp16)[name = string("transpose_3951")]; tensor w_363_cast_fp16 = add(x = transpose_2778, y = transpose_2305)[name = string("w_363_cast_fp16")]; tensor var_1877_cast_fp16 = softmax(axis = var_1685, x = w_363_cast_fp16)[name = string("op_1877_cast_fp16")]; string var_1879_equation_0 = const()[name = string("op_1879_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1879_cast_fp16 = einsum(equation = var_1879_equation_0, values = (var_1775_cast_fp16_10, var_1877_cast_fp16))[name = string("op_1879_cast_fp16")]; tensor transpose_182_perm_0 = const()[name = string("transpose_182_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_914 = const()[name = string("concat_914"), val = tensor([1, 104, 64])]; tensor transpose_182_cast_fp16 = transpose(perm = transpose_182_perm_0, x = var_1741_cast_fp16_11)[name = string("transpose_3950")]; tensor reshape_273_cast_fp16 = reshape(shape = concat_914, x = transpose_182_cast_fp16)[name = string("reshape_273_cast_fp16")]; tensor transpose_183_perm_0 = const()[name = string("transpose_183_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_915 = const()[name = string("concat_915"), val = tensor([1, 64, 104])]; tensor transpose_183_cast_fp16 = transpose(perm = transpose_183_perm_0, x = var_1758_cast_fp16_11)[name = string("transpose_3949")]; tensor reshape_274_cast_fp16 = reshape(shape = concat_915, x = transpose_183_cast_fp16)[name = string("reshape_274_cast_fp16")]; bool matmul_91_transpose_x_0 = const()[name = string("matmul_91_transpose_x_0"), val = bool(false)]; bool matmul_91_transpose_y_0 = const()[name = string("matmul_91_transpose_y_0"), val = bool(false)]; tensor matmul_91_cast_fp16 = matmul(transpose_x = matmul_91_transpose_x_0, transpose_y = matmul_91_transpose_y_0, x = reshape_273_cast_fp16, y = reshape_274_cast_fp16)[name = string("matmul_91_cast_fp16")]; tensor concat_919 = const()[name = string("concat_919"), val = tensor([1, 1, 104, 104])]; tensor reshape_275_cast_fp16 = reshape(shape = concat_919, x = matmul_91_cast_fp16)[name = string("reshape_275_cast_fp16")]; tensor transpose_2779_perm_0 = const()[name = string("transpose_2779_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2779 = transpose(perm = transpose_2779_perm_0, x = reshape_275_cast_fp16)[name = string("transpose_3948")]; tensor w_367_cast_fp16 = add(x = transpose_2779, y = transpose_2305)[name = string("w_367_cast_fp16")]; tensor var_1885_cast_fp16 = softmax(axis = var_1685, x = w_367_cast_fp16)[name = string("op_1885_cast_fp16")]; string var_1887_equation_0 = const()[name = string("op_1887_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1887_cast_fp16 = einsum(equation = var_1887_equation_0, values = (var_1775_cast_fp16_11, var_1885_cast_fp16))[name = string("op_1887_cast_fp16")]; tensor transpose_184_perm_0 = const()[name = string("transpose_184_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_924 = const()[name = string("concat_924"), val = tensor([1, 104, 64])]; tensor transpose_184_cast_fp16 = transpose(perm = transpose_184_perm_0, x = var_1741_cast_fp16_12)[name = string("transpose_3947")]; tensor reshape_276_cast_fp16 = reshape(shape = concat_924, x = transpose_184_cast_fp16)[name = string("reshape_276_cast_fp16")]; tensor transpose_185_perm_0 = const()[name = string("transpose_185_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_925 = const()[name = string("concat_925"), val = tensor([1, 64, 104])]; tensor transpose_185_cast_fp16 = transpose(perm = transpose_185_perm_0, x = var_1758_cast_fp16_12)[name = string("transpose_3946")]; tensor reshape_277_cast_fp16 = reshape(shape = concat_925, x = transpose_185_cast_fp16)[name = string("reshape_277_cast_fp16")]; bool matmul_92_transpose_x_0 = const()[name = string("matmul_92_transpose_x_0"), val = bool(false)]; bool matmul_92_transpose_y_0 = const()[name = string("matmul_92_transpose_y_0"), val = bool(false)]; tensor matmul_92_cast_fp16 = matmul(transpose_x = matmul_92_transpose_x_0, transpose_y = matmul_92_transpose_y_0, x = reshape_276_cast_fp16, y = reshape_277_cast_fp16)[name = string("matmul_92_cast_fp16")]; tensor concat_929 = const()[name = string("concat_929"), val = tensor([1, 1, 104, 104])]; tensor reshape_278_cast_fp16 = reshape(shape = concat_929, x = matmul_92_cast_fp16)[name = string("reshape_278_cast_fp16")]; tensor transpose_2780_perm_0 = const()[name = string("transpose_2780_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2780 = transpose(perm = transpose_2780_perm_0, x = reshape_278_cast_fp16)[name = string("transpose_3945")]; tensor w_371_cast_fp16 = add(x = transpose_2780, y = transpose_2305)[name = string("w_371_cast_fp16")]; tensor var_1893_cast_fp16 = softmax(axis = var_1685, x = w_371_cast_fp16)[name = string("op_1893_cast_fp16")]; string var_1895_equation_0 = const()[name = string("op_1895_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1895_cast_fp16 = einsum(equation = var_1895_equation_0, values = (var_1775_cast_fp16_12, var_1893_cast_fp16))[name = string("op_1895_cast_fp16")]; tensor transpose_186_perm_0 = const()[name = string("transpose_186_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_934 = const()[name = string("concat_934"), val = tensor([1, 104, 64])]; tensor transpose_186_cast_fp16 = transpose(perm = transpose_186_perm_0, x = var_1741_cast_fp16_13)[name = string("transpose_3944")]; tensor reshape_279_cast_fp16 = reshape(shape = concat_934, x = transpose_186_cast_fp16)[name = string("reshape_279_cast_fp16")]; tensor transpose_187_perm_0 = const()[name = string("transpose_187_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_935 = const()[name = string("concat_935"), val = tensor([1, 64, 104])]; tensor transpose_187_cast_fp16 = transpose(perm = transpose_187_perm_0, x = var_1758_cast_fp16_13)[name = string("transpose_3943")]; tensor reshape_280_cast_fp16 = reshape(shape = concat_935, x = transpose_187_cast_fp16)[name = string("reshape_280_cast_fp16")]; bool matmul_93_transpose_x_0 = const()[name = string("matmul_93_transpose_x_0"), val = bool(false)]; bool matmul_93_transpose_y_0 = const()[name = string("matmul_93_transpose_y_0"), val = bool(false)]; tensor matmul_93_cast_fp16 = matmul(transpose_x = matmul_93_transpose_x_0, transpose_y = matmul_93_transpose_y_0, x = reshape_279_cast_fp16, y = reshape_280_cast_fp16)[name = string("matmul_93_cast_fp16")]; tensor concat_939 = const()[name = string("concat_939"), val = tensor([1, 1, 104, 104])]; tensor reshape_281_cast_fp16 = reshape(shape = concat_939, x = matmul_93_cast_fp16)[name = string("reshape_281_cast_fp16")]; tensor transpose_2781_perm_0 = const()[name = string("transpose_2781_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2781 = transpose(perm = transpose_2781_perm_0, x = reshape_281_cast_fp16)[name = string("transpose_3942")]; tensor w_375_cast_fp16 = add(x = transpose_2781, y = transpose_2305)[name = string("w_375_cast_fp16")]; tensor var_1901_cast_fp16 = softmax(axis = var_1685, x = w_375_cast_fp16)[name = string("op_1901_cast_fp16")]; string var_1903_equation_0 = const()[name = string("op_1903_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1903_cast_fp16 = einsum(equation = var_1903_equation_0, values = (var_1775_cast_fp16_13, var_1901_cast_fp16))[name = string("op_1903_cast_fp16")]; tensor transpose_188_perm_0 = const()[name = string("transpose_188_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_944 = const()[name = string("concat_944"), val = tensor([1, 104, 64])]; tensor transpose_188_cast_fp16 = transpose(perm = transpose_188_perm_0, x = var_1741_cast_fp16_14)[name = string("transpose_3941")]; tensor reshape_282_cast_fp16 = reshape(shape = concat_944, x = transpose_188_cast_fp16)[name = string("reshape_282_cast_fp16")]; tensor transpose_189_perm_0 = const()[name = string("transpose_189_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_945 = const()[name = string("concat_945"), val = tensor([1, 64, 104])]; tensor transpose_189_cast_fp16 = transpose(perm = transpose_189_perm_0, x = var_1758_cast_fp16_14)[name = string("transpose_3940")]; tensor reshape_283_cast_fp16 = reshape(shape = concat_945, x = transpose_189_cast_fp16)[name = string("reshape_283_cast_fp16")]; bool matmul_94_transpose_x_0 = const()[name = string("matmul_94_transpose_x_0"), val = bool(false)]; bool matmul_94_transpose_y_0 = const()[name = string("matmul_94_transpose_y_0"), val = bool(false)]; tensor matmul_94_cast_fp16 = matmul(transpose_x = matmul_94_transpose_x_0, transpose_y = matmul_94_transpose_y_0, x = reshape_282_cast_fp16, y = reshape_283_cast_fp16)[name = string("matmul_94_cast_fp16")]; tensor concat_949 = const()[name = string("concat_949"), val = tensor([1, 1, 104, 104])]; tensor reshape_284_cast_fp16 = reshape(shape = concat_949, x = matmul_94_cast_fp16)[name = string("reshape_284_cast_fp16")]; tensor transpose_2782_perm_0 = const()[name = string("transpose_2782_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2782 = transpose(perm = transpose_2782_perm_0, x = reshape_284_cast_fp16)[name = string("transpose_3939")]; tensor w_379_cast_fp16 = add(x = transpose_2782, y = transpose_2305)[name = string("w_379_cast_fp16")]; tensor var_1909_cast_fp16 = softmax(axis = var_1685, x = w_379_cast_fp16)[name = string("op_1909_cast_fp16")]; string var_1911_equation_0 = const()[name = string("op_1911_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1911_cast_fp16 = einsum(equation = var_1911_equation_0, values = (var_1775_cast_fp16_14, var_1909_cast_fp16))[name = string("op_1911_cast_fp16")]; tensor transpose_190_perm_0 = const()[name = string("transpose_190_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_954 = const()[name = string("concat_954"), val = tensor([1, 104, 64])]; tensor transpose_190_cast_fp16 = transpose(perm = transpose_190_perm_0, x = var_1741_cast_fp16_15)[name = string("transpose_3938")]; tensor reshape_285_cast_fp16 = reshape(shape = concat_954, x = transpose_190_cast_fp16)[name = string("reshape_285_cast_fp16")]; tensor transpose_191_perm_0 = const()[name = string("transpose_191_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_955 = const()[name = string("concat_955"), val = tensor([1, 64, 104])]; tensor transpose_191_cast_fp16 = transpose(perm = transpose_191_perm_0, x = var_1758_cast_fp16_15)[name = string("transpose_3937")]; tensor reshape_286_cast_fp16 = reshape(shape = concat_955, x = transpose_191_cast_fp16)[name = string("reshape_286_cast_fp16")]; bool matmul_95_transpose_x_0 = const()[name = string("matmul_95_transpose_x_0"), val = bool(false)]; bool matmul_95_transpose_y_0 = const()[name = string("matmul_95_transpose_y_0"), val = bool(false)]; tensor matmul_95_cast_fp16 = matmul(transpose_x = matmul_95_transpose_x_0, transpose_y = matmul_95_transpose_y_0, x = reshape_285_cast_fp16, y = reshape_286_cast_fp16)[name = string("matmul_95_cast_fp16")]; tensor concat_959 = const()[name = string("concat_959"), val = tensor([1, 1, 104, 104])]; tensor reshape_287_cast_fp16 = reshape(shape = concat_959, x = matmul_95_cast_fp16)[name = string("reshape_287_cast_fp16")]; tensor transpose_2783_perm_0 = const()[name = string("transpose_2783_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2783 = transpose(perm = transpose_2783_perm_0, x = reshape_287_cast_fp16)[name = string("transpose_3936")]; tensor w_383_cast_fp16 = add(x = transpose_2783, y = transpose_2305)[name = string("w_383_cast_fp16")]; tensor var_1917_cast_fp16 = softmax(axis = var_1685, x = w_383_cast_fp16)[name = string("op_1917_cast_fp16")]; string var_1919_equation_0 = const()[name = string("op_1919_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_1919_cast_fp16 = einsum(equation = var_1919_equation_0, values = (var_1775_cast_fp16_15, var_1917_cast_fp16))[name = string("op_1919_cast_fp16")]; bool input_51_interleave_0 = const()[name = string("input_51_interleave_0"), val = bool(false)]; tensor input_51_cast_fp16 = concat(axis = var_1685, interleave = input_51_interleave_0, values = (var_1799_cast_fp16, var_1807_cast_fp16, var_1815_cast_fp16, var_1823_cast_fp16, var_1831_cast_fp16, var_1839_cast_fp16, var_1847_cast_fp16, var_1855_cast_fp16, var_1863_cast_fp16, var_1871_cast_fp16, var_1879_cast_fp16, var_1887_cast_fp16, var_1895_cast_fp16, var_1903_cast_fp16, var_1911_cast_fp16, var_1919_cast_fp16))[name = string("input_51_cast_fp16")]; string var_1928_pad_type_0 = const()[name = string("op_1928_pad_type_0"), val = string("valid")]; tensor var_1928_strides_0 = const()[name = string("op_1928_strides_0"), val = tensor([1, 1])]; tensor var_1928_pad_0 = const()[name = string("op_1928_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_1928_dilations_0 = const()[name = string("op_1928_dilations_0"), val = tensor([1, 1])]; int32 var_1928_groups_0 = const()[name = string("op_1928_groups_0"), val = int32(1)]; tensor layers_5_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_5_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(156337856)))]; tensor layers_5_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_5_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(158435072)))]; tensor var_1928_cast_fp16 = conv(bias = layers_5_self_attn_out_proj_bias_to_fp16, dilations = var_1928_dilations_0, groups = var_1928_groups_0, pad = var_1928_pad_0, pad_type = var_1928_pad_type_0, strides = var_1928_strides_0, weight = layers_5_self_attn_out_proj_weight_to_fp16, x = input_51_cast_fp16)[name = string("op_1928_cast_fp16")]; tensor x_67_cast_fp16 = add(x = x_63_cast_fp16, y = var_1928_cast_fp16)[name = string("x_67_cast_fp16")]; tensor mu_23_axes_0 = const()[name = string("mu_23_axes_0"), val = tensor([1])]; bool mu_23_keep_dims_0 = const()[name = string("mu_23_keep_dims_0"), val = bool(true)]; tensor mu_23_cast_fp16 = reduce_mean(axes = mu_23_axes_0, keep_dims = mu_23_keep_dims_0, x = x_67_cast_fp16)[name = string("mu_23_cast_fp16")]; tensor var_1934_cast_fp16 = sub(x = x_67_cast_fp16, y = mu_23_cast_fp16)[name = string("op_1934_cast_fp16")]; fp16 var_1688_promoted_1_to_fp16 = const()[name = string("op_1688_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_1935_cast_fp16 = pow(x = var_1934_cast_fp16, y = var_1688_promoted_1_to_fp16)[name = string("op_1935_cast_fp16")]; tensor var_23_axes_0 = const()[name = string("var_23_axes_0"), val = tensor([1])]; bool var_23_keep_dims_0 = const()[name = string("var_23_keep_dims_0"), val = bool(true)]; tensor var_23_cast_fp16 = reduce_mean(axes = var_23_axes_0, keep_dims = var_23_keep_dims_0, x = var_1935_cast_fp16)[name = string("var_23_cast_fp16")]; fp16 var_1939_to_fp16 = const()[name = string("op_1939_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_1940_cast_fp16 = add(x = var_23_cast_fp16, y = var_1939_to_fp16)[name = string("op_1940_cast_fp16")]; fp32 var_1941_epsilon_0 = const()[name = string("op_1941_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_1941_cast_fp16 = rsqrt(epsilon = var_1941_epsilon_0, x = var_1940_cast_fp16)[name = string("op_1941_cast_fp16")]; tensor x_69_cast_fp16 = mul(x = var_1934_cast_fp16, y = var_1941_cast_fp16)[name = string("x_69_cast_fp16")]; tensor input_53_gamma_0_to_fp16 = const()[name = string("input_53_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(158437184)))]; tensor input_53_beta_0_to_fp16 = const()[name = string("input_53_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(158439296)))]; fp16 input_53_epsilon_0_to_fp16 = const()[name = string("input_53_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_53_cast_fp16 = batch_norm(beta = input_53_beta_0_to_fp16, epsilon = input_53_epsilon_0_to_fp16, gamma = input_53_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_69_cast_fp16)[name = string("input_53_cast_fp16")]; string x_71_pad_type_0 = const()[name = string("x_71_pad_type_0"), val = string("valid")]; tensor x_71_strides_0 = const()[name = string("x_71_strides_0"), val = tensor([1, 1])]; tensor x_71_pad_0 = const()[name = string("x_71_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_71_dilations_0 = const()[name = string("x_71_dilations_0"), val = tensor([1, 1])]; int32 x_71_groups_0 = const()[name = string("x_71_groups_0"), val = int32(1)]; tensor layers_5_fc1_weight_to_fp16 = const()[name = string("layers_5_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(158441408)))]; tensor layers_5_fc1_bias_to_fp16 = const()[name = string("layers_5_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(166830080)))]; tensor x_71_cast_fp16 = conv(bias = layers_5_fc1_bias_to_fp16, dilations = x_71_dilations_0, groups = x_71_groups_0, pad = x_71_pad_0, pad_type = x_71_pad_type_0, strides = x_71_strides_0, weight = layers_5_fc1_weight_to_fp16, x = input_53_cast_fp16)[name = string("x_71_cast_fp16")]; fp16 var_1956_to_fp16 = const()[name = string("op_1956_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_1957_cast_fp16 = mul(x = x_71_cast_fp16, y = var_1956_to_fp16)[name = string("op_1957_cast_fp16")]; tensor var_1958_cast_fp16 = mul(x = var_1957_cast_fp16, y = x_71_cast_fp16)[name = string("op_1958_cast_fp16")]; tensor var_1959_cast_fp16 = mul(x = var_1958_cast_fp16, y = x_71_cast_fp16)[name = string("op_1959_cast_fp16")]; tensor var_1960_cast_fp16 = add(x = x_71_cast_fp16, y = var_1959_cast_fp16)[name = string("op_1960_cast_fp16")]; fp16 var_1961_to_fp16 = const()[name = string("op_1961_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_17_cast_fp16 = mul(x = var_1960_cast_fp16, y = var_1961_to_fp16)[name = string("u_17_cast_fp16")]; fp16 var_1963_to_fp16 = const()[name = string("op_1963_to_fp16"), val = fp16(0x1p-1)]; tensor var_1964_cast_fp16 = mul(x = x_71_cast_fp16, y = var_1963_to_fp16)[name = string("op_1964_cast_fp16")]; tensor var_1965_cast_fp16 = tanh(x = u_17_cast_fp16)[name = string("op_1965_cast_fp16")]; fp16 var_1966_to_fp16 = const()[name = string("op_1966_to_fp16"), val = fp16(0x1p+0)]; tensor var_1967_cast_fp16 = add(x = var_1965_cast_fp16, y = var_1966_to_fp16)[name = string("op_1967_cast_fp16")]; tensor input_55_cast_fp16 = mul(x = var_1964_cast_fp16, y = var_1967_cast_fp16)[name = string("input_55_cast_fp16")]; string h_11_pad_type_0 = const()[name = string("h_11_pad_type_0"), val = string("valid")]; tensor h_11_strides_0 = const()[name = string("h_11_strides_0"), val = tensor([1, 1])]; tensor h_11_pad_0 = const()[name = string("h_11_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_11_dilations_0 = const()[name = string("h_11_dilations_0"), val = tensor([1, 1])]; int32 h_11_groups_0 = const()[name = string("h_11_groups_0"), val = int32(1)]; tensor layers_5_fc2_weight_to_fp16 = const()[name = string("layers_5_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(166838336)))]; tensor layers_5_fc2_bias_to_fp16 = const()[name = string("layers_5_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(175227008)))]; tensor h_11_cast_fp16 = conv(bias = layers_5_fc2_bias_to_fp16, dilations = h_11_dilations_0, groups = h_11_groups_0, pad = h_11_pad_0, pad_type = h_11_pad_type_0, strides = h_11_strides_0, weight = layers_5_fc2_weight_to_fp16, x = input_55_cast_fp16)[name = string("h_11_cast_fp16")]; tensor x_73_cast_fp16 = add(x = x_67_cast_fp16, y = h_11_cast_fp16)[name = string("x_73_cast_fp16")]; int32 var_1983 = const()[name = string("op_1983"), val = int32(1)]; tensor mu_25_axes_0 = const()[name = string("mu_25_axes_0"), val = tensor([1])]; bool mu_25_keep_dims_0 = const()[name = string("mu_25_keep_dims_0"), val = bool(true)]; tensor mu_25_cast_fp16 = reduce_mean(axes = mu_25_axes_0, keep_dims = mu_25_keep_dims_0, x = x_73_cast_fp16)[name = string("mu_25_cast_fp16")]; tensor var_1997_cast_fp16 = sub(x = x_73_cast_fp16, y = mu_25_cast_fp16)[name = string("op_1997_cast_fp16")]; fp16 var_1986_promoted_to_fp16 = const()[name = string("op_1986_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_1998_cast_fp16 = pow(x = var_1997_cast_fp16, y = var_1986_promoted_to_fp16)[name = string("op_1998_cast_fp16")]; tensor var_25_axes_0 = const()[name = string("var_25_axes_0"), val = tensor([1])]; bool var_25_keep_dims_0 = const()[name = string("var_25_keep_dims_0"), val = bool(true)]; tensor var_25_cast_fp16 = reduce_mean(axes = var_25_axes_0, keep_dims = var_25_keep_dims_0, x = var_1998_cast_fp16)[name = string("var_25_cast_fp16")]; fp16 var_2002_to_fp16 = const()[name = string("op_2002_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2003_cast_fp16 = add(x = var_25_cast_fp16, y = var_2002_to_fp16)[name = string("op_2003_cast_fp16")]; fp32 var_2004_epsilon_0 = const()[name = string("op_2004_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2004_cast_fp16 = rsqrt(epsilon = var_2004_epsilon_0, x = var_2003_cast_fp16)[name = string("op_2004_cast_fp16")]; tensor x_75_cast_fp16 = mul(x = var_1997_cast_fp16, y = var_2004_cast_fp16)[name = string("x_75_cast_fp16")]; tensor input_57_gamma_0_to_fp16 = const()[name = string("input_57_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(175229120)))]; tensor input_57_beta_0_to_fp16 = const()[name = string("input_57_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(175231232)))]; fp16 input_57_epsilon_0_to_fp16 = const()[name = string("input_57_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_57_cast_fp16 = batch_norm(beta = input_57_beta_0_to_fp16, epsilon = input_57_epsilon_0_to_fp16, gamma = input_57_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_75_cast_fp16)[name = string("input_57_cast_fp16")]; string var_2022_pad_type_0 = const()[name = string("op_2022_pad_type_0"), val = string("valid")]; tensor var_2022_strides_0 = const()[name = string("op_2022_strides_0"), val = tensor([1, 1])]; tensor var_2022_pad_0 = const()[name = string("op_2022_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2022_dilations_0 = const()[name = string("op_2022_dilations_0"), val = tensor([1, 1])]; int32 var_2022_groups_0 = const()[name = string("op_2022_groups_0"), val = int32(1)]; tensor var_2024_weight_0_to_fp16 = const()[name = string("op_2024_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(175233344)))]; tensor var_2024_bias_0_to_fp16 = const()[name = string("op_2024_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(177330560)))]; tensor var_2024_cast_fp16 = conv(bias = var_2024_bias_0_to_fp16, dilations = var_2022_dilations_0, groups = var_2022_groups_0, pad = var_2022_pad_0, pad_type = var_2022_pad_type_0, strides = var_2022_strides_0, weight = var_2024_weight_0_to_fp16, x = input_57_cast_fp16)[name = string("op_2024_cast_fp16")]; string var_2031_pad_type_0 = const()[name = string("op_2031_pad_type_0"), val = string("valid")]; tensor var_2031_strides_0 = const()[name = string("op_2031_strides_0"), val = tensor([1, 1])]; tensor var_2031_pad_0 = const()[name = string("op_2031_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2031_dilations_0 = const()[name = string("op_2031_dilations_0"), val = tensor([1, 1])]; int32 var_2031_groups_0 = const()[name = string("op_2031_groups_0"), val = int32(1)]; tensor layers_6_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_6_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(177332672)))]; tensor layers_6_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_6_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(179429888)))]; tensor var_2031_cast_fp16 = conv(bias = layers_6_self_attn_k_proj_bias_to_fp16, dilations = var_2031_dilations_0, groups = var_2031_groups_0, pad = var_2031_pad_0, pad_type = var_2031_pad_type_0, strides = var_2031_strides_0, weight = layers_6_self_attn_k_proj_weight_to_fp16, x = input_57_cast_fp16)[name = string("op_2031_cast_fp16")]; string var_2038_pad_type_0 = const()[name = string("op_2038_pad_type_0"), val = string("valid")]; tensor var_2038_strides_0 = const()[name = string("op_2038_strides_0"), val = tensor([1, 1])]; tensor var_2038_pad_0 = const()[name = string("op_2038_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2038_dilations_0 = const()[name = string("op_2038_dilations_0"), val = tensor([1, 1])]; int32 var_2038_groups_0 = const()[name = string("op_2038_groups_0"), val = int32(1)]; tensor layers_6_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_6_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(179432000)))]; tensor layers_6_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_6_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(181529216)))]; tensor var_2038_cast_fp16 = conv(bias = layers_6_self_attn_v_proj_bias_to_fp16, dilations = var_2038_dilations_0, groups = var_2038_groups_0, pad = var_2038_pad_0, pad_type = var_2038_pad_type_0, strides = var_2038_strides_0, weight = layers_6_self_attn_v_proj_weight_to_fp16, x = input_57_cast_fp16)[name = string("op_2038_cast_fp16")]; tensor tile_18 = const()[name = string("tile_18"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(181531328)))]; int32 var_2039_axis_0 = const()[name = string("op_2039_axis_0"), val = int32(1)]; tensor var_2039_cast_fp16_0, tensor var_2039_cast_fp16_1, tensor var_2039_cast_fp16_2, tensor var_2039_cast_fp16_3, tensor var_2039_cast_fp16_4, tensor var_2039_cast_fp16_5, tensor var_2039_cast_fp16_6, tensor var_2039_cast_fp16_7, tensor var_2039_cast_fp16_8, tensor var_2039_cast_fp16_9, tensor var_2039_cast_fp16_10, tensor var_2039_cast_fp16_11, tensor var_2039_cast_fp16_12, tensor var_2039_cast_fp16_13, tensor var_2039_cast_fp16_14, tensor var_2039_cast_fp16_15 = split(axis = var_2039_axis_0, split_sizes = tile_18, x = var_2024_cast_fp16)[name = string("op_2039_cast_fp16")]; tensor tile_19 = const()[name = string("tile_19"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(181531456)))]; int32 var_2056_axis_0 = const()[name = string("op_2056_axis_0"), val = int32(1)]; tensor var_2056_cast_fp16_0, tensor var_2056_cast_fp16_1, tensor var_2056_cast_fp16_2, tensor var_2056_cast_fp16_3, tensor var_2056_cast_fp16_4, tensor var_2056_cast_fp16_5, tensor var_2056_cast_fp16_6, tensor var_2056_cast_fp16_7, tensor var_2056_cast_fp16_8, tensor var_2056_cast_fp16_9, tensor var_2056_cast_fp16_10, tensor var_2056_cast_fp16_11, tensor var_2056_cast_fp16_12, tensor var_2056_cast_fp16_13, tensor var_2056_cast_fp16_14, tensor var_2056_cast_fp16_15 = split(axis = var_2056_axis_0, split_sizes = tile_19, x = var_2031_cast_fp16)[name = string("op_2056_cast_fp16")]; tensor tile_20 = const()[name = string("tile_20"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(181531584)))]; int32 var_2073_axis_0 = const()[name = string("op_2073_axis_0"), val = int32(1)]; tensor var_2073_cast_fp16_0, tensor var_2073_cast_fp16_1, tensor var_2073_cast_fp16_2, tensor var_2073_cast_fp16_3, tensor var_2073_cast_fp16_4, tensor var_2073_cast_fp16_5, tensor var_2073_cast_fp16_6, tensor var_2073_cast_fp16_7, tensor var_2073_cast_fp16_8, tensor var_2073_cast_fp16_9, tensor var_2073_cast_fp16_10, tensor var_2073_cast_fp16_11, tensor var_2073_cast_fp16_12, tensor var_2073_cast_fp16_13, tensor var_2073_cast_fp16_14, tensor var_2073_cast_fp16_15 = split(axis = var_2073_axis_0, split_sizes = tile_20, x = var_2038_cast_fp16)[name = string("op_2073_cast_fp16")]; tensor transpose_192_perm_0 = const()[name = string("transpose_192_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_964 = const()[name = string("concat_964"), val = tensor([1, 104, 64])]; tensor transpose_192_cast_fp16 = transpose(perm = transpose_192_perm_0, x = var_2039_cast_fp16_0)[name = string("transpose_3935")]; tensor reshape_288_cast_fp16 = reshape(shape = concat_964, x = transpose_192_cast_fp16)[name = string("reshape_288_cast_fp16")]; tensor transpose_193_perm_0 = const()[name = string("transpose_193_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_965 = const()[name = string("concat_965"), val = tensor([1, 64, 104])]; tensor transpose_193_cast_fp16 = transpose(perm = transpose_193_perm_0, x = var_2056_cast_fp16_0)[name = string("transpose_3934")]; tensor reshape_289_cast_fp16 = reshape(shape = concat_965, x = transpose_193_cast_fp16)[name = string("reshape_289_cast_fp16")]; bool matmul_96_transpose_x_0 = const()[name = string("matmul_96_transpose_x_0"), val = bool(false)]; bool matmul_96_transpose_y_0 = const()[name = string("matmul_96_transpose_y_0"), val = bool(false)]; tensor matmul_96_cast_fp16 = matmul(transpose_x = matmul_96_transpose_x_0, transpose_y = matmul_96_transpose_y_0, x = reshape_288_cast_fp16, y = reshape_289_cast_fp16)[name = string("matmul_96_cast_fp16")]; tensor concat_969 = const()[name = string("concat_969"), val = tensor([1, 1, 104, 104])]; tensor reshape_290_cast_fp16 = reshape(shape = concat_969, x = matmul_96_cast_fp16)[name = string("reshape_290_cast_fp16")]; tensor transpose_2784_perm_0 = const()[name = string("transpose_2784_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2784 = transpose(perm = transpose_2784_perm_0, x = reshape_290_cast_fp16)[name = string("transpose_3933")]; tensor w_387_cast_fp16 = add(x = transpose_2784, y = transpose_2305)[name = string("w_387_cast_fp16")]; tensor var_2095_cast_fp16 = softmax(axis = var_1983, x = w_387_cast_fp16)[name = string("op_2095_cast_fp16")]; string var_2097_equation_0 = const()[name = string("op_2097_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2097_cast_fp16 = einsum(equation = var_2097_equation_0, values = (var_2073_cast_fp16_0, var_2095_cast_fp16))[name = string("op_2097_cast_fp16")]; tensor transpose_194_perm_0 = const()[name = string("transpose_194_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_974 = const()[name = string("concat_974"), val = tensor([1, 104, 64])]; tensor transpose_194_cast_fp16 = transpose(perm = transpose_194_perm_0, x = var_2039_cast_fp16_1)[name = string("transpose_3932")]; tensor reshape_291_cast_fp16 = reshape(shape = concat_974, x = transpose_194_cast_fp16)[name = string("reshape_291_cast_fp16")]; tensor transpose_195_perm_0 = const()[name = string("transpose_195_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_975 = const()[name = string("concat_975"), val = tensor([1, 64, 104])]; tensor transpose_195_cast_fp16 = transpose(perm = transpose_195_perm_0, x = var_2056_cast_fp16_1)[name = string("transpose_3931")]; tensor reshape_292_cast_fp16 = reshape(shape = concat_975, x = transpose_195_cast_fp16)[name = string("reshape_292_cast_fp16")]; bool matmul_97_transpose_x_0 = const()[name = string("matmul_97_transpose_x_0"), val = bool(false)]; bool matmul_97_transpose_y_0 = const()[name = string("matmul_97_transpose_y_0"), val = bool(false)]; tensor matmul_97_cast_fp16 = matmul(transpose_x = matmul_97_transpose_x_0, transpose_y = matmul_97_transpose_y_0, x = reshape_291_cast_fp16, y = reshape_292_cast_fp16)[name = string("matmul_97_cast_fp16")]; tensor concat_979 = const()[name = string("concat_979"), val = tensor([1, 1, 104, 104])]; tensor reshape_293_cast_fp16 = reshape(shape = concat_979, x = matmul_97_cast_fp16)[name = string("reshape_293_cast_fp16")]; tensor transpose_2785_perm_0 = const()[name = string("transpose_2785_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2785 = transpose(perm = transpose_2785_perm_0, x = reshape_293_cast_fp16)[name = string("transpose_3930")]; tensor w_391_cast_fp16 = add(x = transpose_2785, y = transpose_2305)[name = string("w_391_cast_fp16")]; tensor var_2103_cast_fp16 = softmax(axis = var_1983, x = w_391_cast_fp16)[name = string("op_2103_cast_fp16")]; string var_2105_equation_0 = const()[name = string("op_2105_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2105_cast_fp16 = einsum(equation = var_2105_equation_0, values = (var_2073_cast_fp16_1, var_2103_cast_fp16))[name = string("op_2105_cast_fp16")]; tensor transpose_196_perm_0 = const()[name = string("transpose_196_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_984 = const()[name = string("concat_984"), val = tensor([1, 104, 64])]; tensor transpose_196_cast_fp16 = transpose(perm = transpose_196_perm_0, x = var_2039_cast_fp16_2)[name = string("transpose_3929")]; tensor reshape_294_cast_fp16 = reshape(shape = concat_984, x = transpose_196_cast_fp16)[name = string("reshape_294_cast_fp16")]; tensor transpose_197_perm_0 = const()[name = string("transpose_197_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_985 = const()[name = string("concat_985"), val = tensor([1, 64, 104])]; tensor transpose_197_cast_fp16 = transpose(perm = transpose_197_perm_0, x = var_2056_cast_fp16_2)[name = string("transpose_3928")]; tensor reshape_295_cast_fp16 = reshape(shape = concat_985, x = transpose_197_cast_fp16)[name = string("reshape_295_cast_fp16")]; bool matmul_98_transpose_x_0 = const()[name = string("matmul_98_transpose_x_0"), val = bool(false)]; bool matmul_98_transpose_y_0 = const()[name = string("matmul_98_transpose_y_0"), val = bool(false)]; tensor matmul_98_cast_fp16 = matmul(transpose_x = matmul_98_transpose_x_0, transpose_y = matmul_98_transpose_y_0, x = reshape_294_cast_fp16, y = reshape_295_cast_fp16)[name = string("matmul_98_cast_fp16")]; tensor concat_989 = const()[name = string("concat_989"), val = tensor([1, 1, 104, 104])]; tensor reshape_296_cast_fp16 = reshape(shape = concat_989, x = matmul_98_cast_fp16)[name = string("reshape_296_cast_fp16")]; tensor transpose_2786_perm_0 = const()[name = string("transpose_2786_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2786 = transpose(perm = transpose_2786_perm_0, x = reshape_296_cast_fp16)[name = string("transpose_3927")]; tensor w_395_cast_fp16 = add(x = transpose_2786, y = transpose_2305)[name = string("w_395_cast_fp16")]; tensor var_2111_cast_fp16 = softmax(axis = var_1983, x = w_395_cast_fp16)[name = string("op_2111_cast_fp16")]; string var_2113_equation_0 = const()[name = string("op_2113_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2113_cast_fp16 = einsum(equation = var_2113_equation_0, values = (var_2073_cast_fp16_2, var_2111_cast_fp16))[name = string("op_2113_cast_fp16")]; tensor transpose_198_perm_0 = const()[name = string("transpose_198_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_994 = const()[name = string("concat_994"), val = tensor([1, 104, 64])]; tensor transpose_198_cast_fp16 = transpose(perm = transpose_198_perm_0, x = var_2039_cast_fp16_3)[name = string("transpose_3926")]; tensor reshape_297_cast_fp16 = reshape(shape = concat_994, x = transpose_198_cast_fp16)[name = string("reshape_297_cast_fp16")]; tensor transpose_199_perm_0 = const()[name = string("transpose_199_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_995 = const()[name = string("concat_995"), val = tensor([1, 64, 104])]; tensor transpose_199_cast_fp16 = transpose(perm = transpose_199_perm_0, x = var_2056_cast_fp16_3)[name = string("transpose_3925")]; tensor reshape_298_cast_fp16 = reshape(shape = concat_995, x = transpose_199_cast_fp16)[name = string("reshape_298_cast_fp16")]; bool matmul_99_transpose_x_0 = const()[name = string("matmul_99_transpose_x_0"), val = bool(false)]; bool matmul_99_transpose_y_0 = const()[name = string("matmul_99_transpose_y_0"), val = bool(false)]; tensor matmul_99_cast_fp16 = matmul(transpose_x = matmul_99_transpose_x_0, transpose_y = matmul_99_transpose_y_0, x = reshape_297_cast_fp16, y = reshape_298_cast_fp16)[name = string("matmul_99_cast_fp16")]; tensor concat_999 = const()[name = string("concat_999"), val = tensor([1, 1, 104, 104])]; tensor reshape_299_cast_fp16 = reshape(shape = concat_999, x = matmul_99_cast_fp16)[name = string("reshape_299_cast_fp16")]; tensor transpose_2787_perm_0 = const()[name = string("transpose_2787_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2787 = transpose(perm = transpose_2787_perm_0, x = reshape_299_cast_fp16)[name = string("transpose_3924")]; tensor w_399_cast_fp16 = add(x = transpose_2787, y = transpose_2305)[name = string("w_399_cast_fp16")]; tensor var_2119_cast_fp16 = softmax(axis = var_1983, x = w_399_cast_fp16)[name = string("op_2119_cast_fp16")]; string var_2121_equation_0 = const()[name = string("op_2121_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2121_cast_fp16 = einsum(equation = var_2121_equation_0, values = (var_2073_cast_fp16_3, var_2119_cast_fp16))[name = string("op_2121_cast_fp16")]; tensor transpose_200_perm_0 = const()[name = string("transpose_200_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1004 = const()[name = string("concat_1004"), val = tensor([1, 104, 64])]; tensor transpose_200_cast_fp16 = transpose(perm = transpose_200_perm_0, x = var_2039_cast_fp16_4)[name = string("transpose_3923")]; tensor reshape_300_cast_fp16 = reshape(shape = concat_1004, x = transpose_200_cast_fp16)[name = string("reshape_300_cast_fp16")]; tensor transpose_201_perm_0 = const()[name = string("transpose_201_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1005 = const()[name = string("concat_1005"), val = tensor([1, 64, 104])]; tensor transpose_201_cast_fp16 = transpose(perm = transpose_201_perm_0, x = var_2056_cast_fp16_4)[name = string("transpose_3922")]; tensor reshape_301_cast_fp16 = reshape(shape = concat_1005, x = transpose_201_cast_fp16)[name = string("reshape_301_cast_fp16")]; bool matmul_100_transpose_x_0 = const()[name = string("matmul_100_transpose_x_0"), val = bool(false)]; bool matmul_100_transpose_y_0 = const()[name = string("matmul_100_transpose_y_0"), val = bool(false)]; tensor matmul_100_cast_fp16 = matmul(transpose_x = matmul_100_transpose_x_0, transpose_y = matmul_100_transpose_y_0, x = reshape_300_cast_fp16, y = reshape_301_cast_fp16)[name = string("matmul_100_cast_fp16")]; tensor concat_1009 = const()[name = string("concat_1009"), val = tensor([1, 1, 104, 104])]; tensor reshape_302_cast_fp16 = reshape(shape = concat_1009, x = matmul_100_cast_fp16)[name = string("reshape_302_cast_fp16")]; tensor transpose_2788_perm_0 = const()[name = string("transpose_2788_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2788 = transpose(perm = transpose_2788_perm_0, x = reshape_302_cast_fp16)[name = string("transpose_3921")]; tensor w_403_cast_fp16 = add(x = transpose_2788, y = transpose_2305)[name = string("w_403_cast_fp16")]; tensor var_2127_cast_fp16 = softmax(axis = var_1983, x = w_403_cast_fp16)[name = string("op_2127_cast_fp16")]; string var_2129_equation_0 = const()[name = string("op_2129_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2129_cast_fp16 = einsum(equation = var_2129_equation_0, values = (var_2073_cast_fp16_4, var_2127_cast_fp16))[name = string("op_2129_cast_fp16")]; tensor transpose_202_perm_0 = const()[name = string("transpose_202_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1014 = const()[name = string("concat_1014"), val = tensor([1, 104, 64])]; tensor transpose_202_cast_fp16 = transpose(perm = transpose_202_perm_0, x = var_2039_cast_fp16_5)[name = string("transpose_3920")]; tensor reshape_303_cast_fp16 = reshape(shape = concat_1014, x = transpose_202_cast_fp16)[name = string("reshape_303_cast_fp16")]; tensor transpose_203_perm_0 = const()[name = string("transpose_203_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1015 = const()[name = string("concat_1015"), val = tensor([1, 64, 104])]; tensor transpose_203_cast_fp16 = transpose(perm = transpose_203_perm_0, x = var_2056_cast_fp16_5)[name = string("transpose_3919")]; tensor reshape_304_cast_fp16 = reshape(shape = concat_1015, x = transpose_203_cast_fp16)[name = string("reshape_304_cast_fp16")]; bool matmul_101_transpose_x_0 = const()[name = string("matmul_101_transpose_x_0"), val = bool(false)]; bool matmul_101_transpose_y_0 = const()[name = string("matmul_101_transpose_y_0"), val = bool(false)]; tensor matmul_101_cast_fp16 = matmul(transpose_x = matmul_101_transpose_x_0, transpose_y = matmul_101_transpose_y_0, x = reshape_303_cast_fp16, y = reshape_304_cast_fp16)[name = string("matmul_101_cast_fp16")]; tensor concat_1019 = const()[name = string("concat_1019"), val = tensor([1, 1, 104, 104])]; tensor reshape_305_cast_fp16 = reshape(shape = concat_1019, x = matmul_101_cast_fp16)[name = string("reshape_305_cast_fp16")]; tensor transpose_2789_perm_0 = const()[name = string("transpose_2789_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2789 = transpose(perm = transpose_2789_perm_0, x = reshape_305_cast_fp16)[name = string("transpose_3918")]; tensor w_407_cast_fp16 = add(x = transpose_2789, y = transpose_2305)[name = string("w_407_cast_fp16")]; tensor var_2135_cast_fp16 = softmax(axis = var_1983, x = w_407_cast_fp16)[name = string("op_2135_cast_fp16")]; string var_2137_equation_0 = const()[name = string("op_2137_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2137_cast_fp16 = einsum(equation = var_2137_equation_0, values = (var_2073_cast_fp16_5, var_2135_cast_fp16))[name = string("op_2137_cast_fp16")]; tensor transpose_204_perm_0 = const()[name = string("transpose_204_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1024 = const()[name = string("concat_1024"), val = tensor([1, 104, 64])]; tensor transpose_204_cast_fp16 = transpose(perm = transpose_204_perm_0, x = var_2039_cast_fp16_6)[name = string("transpose_3917")]; tensor reshape_306_cast_fp16 = reshape(shape = concat_1024, x = transpose_204_cast_fp16)[name = string("reshape_306_cast_fp16")]; tensor transpose_205_perm_0 = const()[name = string("transpose_205_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1025 = const()[name = string("concat_1025"), val = tensor([1, 64, 104])]; tensor transpose_205_cast_fp16 = transpose(perm = transpose_205_perm_0, x = var_2056_cast_fp16_6)[name = string("transpose_3916")]; tensor reshape_307_cast_fp16 = reshape(shape = concat_1025, x = transpose_205_cast_fp16)[name = string("reshape_307_cast_fp16")]; bool matmul_102_transpose_x_0 = const()[name = string("matmul_102_transpose_x_0"), val = bool(false)]; bool matmul_102_transpose_y_0 = const()[name = string("matmul_102_transpose_y_0"), val = bool(false)]; tensor matmul_102_cast_fp16 = matmul(transpose_x = matmul_102_transpose_x_0, transpose_y = matmul_102_transpose_y_0, x = reshape_306_cast_fp16, y = reshape_307_cast_fp16)[name = string("matmul_102_cast_fp16")]; tensor concat_1029 = const()[name = string("concat_1029"), val = tensor([1, 1, 104, 104])]; tensor reshape_308_cast_fp16 = reshape(shape = concat_1029, x = matmul_102_cast_fp16)[name = string("reshape_308_cast_fp16")]; tensor transpose_2790_perm_0 = const()[name = string("transpose_2790_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2790 = transpose(perm = transpose_2790_perm_0, x = reshape_308_cast_fp16)[name = string("transpose_3915")]; tensor w_411_cast_fp16 = add(x = transpose_2790, y = transpose_2305)[name = string("w_411_cast_fp16")]; tensor var_2143_cast_fp16 = softmax(axis = var_1983, x = w_411_cast_fp16)[name = string("op_2143_cast_fp16")]; string var_2145_equation_0 = const()[name = string("op_2145_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2145_cast_fp16 = einsum(equation = var_2145_equation_0, values = (var_2073_cast_fp16_6, var_2143_cast_fp16))[name = string("op_2145_cast_fp16")]; tensor transpose_206_perm_0 = const()[name = string("transpose_206_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1034 = const()[name = string("concat_1034"), val = tensor([1, 104, 64])]; tensor transpose_206_cast_fp16 = transpose(perm = transpose_206_perm_0, x = var_2039_cast_fp16_7)[name = string("transpose_3914")]; tensor reshape_309_cast_fp16 = reshape(shape = concat_1034, x = transpose_206_cast_fp16)[name = string("reshape_309_cast_fp16")]; tensor transpose_207_perm_0 = const()[name = string("transpose_207_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1035 = const()[name = string("concat_1035"), val = tensor([1, 64, 104])]; tensor transpose_207_cast_fp16 = transpose(perm = transpose_207_perm_0, x = var_2056_cast_fp16_7)[name = string("transpose_3913")]; tensor reshape_310_cast_fp16 = reshape(shape = concat_1035, x = transpose_207_cast_fp16)[name = string("reshape_310_cast_fp16")]; bool matmul_103_transpose_x_0 = const()[name = string("matmul_103_transpose_x_0"), val = bool(false)]; bool matmul_103_transpose_y_0 = const()[name = string("matmul_103_transpose_y_0"), val = bool(false)]; tensor matmul_103_cast_fp16 = matmul(transpose_x = matmul_103_transpose_x_0, transpose_y = matmul_103_transpose_y_0, x = reshape_309_cast_fp16, y = reshape_310_cast_fp16)[name = string("matmul_103_cast_fp16")]; tensor concat_1039 = const()[name = string("concat_1039"), val = tensor([1, 1, 104, 104])]; tensor reshape_311_cast_fp16 = reshape(shape = concat_1039, x = matmul_103_cast_fp16)[name = string("reshape_311_cast_fp16")]; tensor transpose_2791_perm_0 = const()[name = string("transpose_2791_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2791 = transpose(perm = transpose_2791_perm_0, x = reshape_311_cast_fp16)[name = string("transpose_3912")]; tensor w_415_cast_fp16 = add(x = transpose_2791, y = transpose_2305)[name = string("w_415_cast_fp16")]; tensor var_2151_cast_fp16 = softmax(axis = var_1983, x = w_415_cast_fp16)[name = string("op_2151_cast_fp16")]; string var_2153_equation_0 = const()[name = string("op_2153_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2153_cast_fp16 = einsum(equation = var_2153_equation_0, values = (var_2073_cast_fp16_7, var_2151_cast_fp16))[name = string("op_2153_cast_fp16")]; tensor transpose_208_perm_0 = const()[name = string("transpose_208_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1044 = const()[name = string("concat_1044"), val = tensor([1, 104, 64])]; tensor transpose_208_cast_fp16 = transpose(perm = transpose_208_perm_0, x = var_2039_cast_fp16_8)[name = string("transpose_3911")]; tensor reshape_312_cast_fp16 = reshape(shape = concat_1044, x = transpose_208_cast_fp16)[name = string("reshape_312_cast_fp16")]; tensor transpose_209_perm_0 = const()[name = string("transpose_209_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1045 = const()[name = string("concat_1045"), val = tensor([1, 64, 104])]; tensor transpose_209_cast_fp16 = transpose(perm = transpose_209_perm_0, x = var_2056_cast_fp16_8)[name = string("transpose_3910")]; tensor reshape_313_cast_fp16 = reshape(shape = concat_1045, x = transpose_209_cast_fp16)[name = string("reshape_313_cast_fp16")]; bool matmul_104_transpose_x_0 = const()[name = string("matmul_104_transpose_x_0"), val = bool(false)]; bool matmul_104_transpose_y_0 = const()[name = string("matmul_104_transpose_y_0"), val = bool(false)]; tensor matmul_104_cast_fp16 = matmul(transpose_x = matmul_104_transpose_x_0, transpose_y = matmul_104_transpose_y_0, x = reshape_312_cast_fp16, y = reshape_313_cast_fp16)[name = string("matmul_104_cast_fp16")]; tensor concat_1049 = const()[name = string("concat_1049"), val = tensor([1, 1, 104, 104])]; tensor reshape_314_cast_fp16 = reshape(shape = concat_1049, x = matmul_104_cast_fp16)[name = string("reshape_314_cast_fp16")]; tensor transpose_2792_perm_0 = const()[name = string("transpose_2792_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2792 = transpose(perm = transpose_2792_perm_0, x = reshape_314_cast_fp16)[name = string("transpose_3909")]; tensor w_419_cast_fp16 = add(x = transpose_2792, y = transpose_2305)[name = string("w_419_cast_fp16")]; tensor var_2159_cast_fp16 = softmax(axis = var_1983, x = w_419_cast_fp16)[name = string("op_2159_cast_fp16")]; string var_2161_equation_0 = const()[name = string("op_2161_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2161_cast_fp16 = einsum(equation = var_2161_equation_0, values = (var_2073_cast_fp16_8, var_2159_cast_fp16))[name = string("op_2161_cast_fp16")]; tensor transpose_210_perm_0 = const()[name = string("transpose_210_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1054 = const()[name = string("concat_1054"), val = tensor([1, 104, 64])]; tensor transpose_210_cast_fp16 = transpose(perm = transpose_210_perm_0, x = var_2039_cast_fp16_9)[name = string("transpose_3908")]; tensor reshape_315_cast_fp16 = reshape(shape = concat_1054, x = transpose_210_cast_fp16)[name = string("reshape_315_cast_fp16")]; tensor transpose_211_perm_0 = const()[name = string("transpose_211_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1055 = const()[name = string("concat_1055"), val = tensor([1, 64, 104])]; tensor transpose_211_cast_fp16 = transpose(perm = transpose_211_perm_0, x = var_2056_cast_fp16_9)[name = string("transpose_3907")]; tensor reshape_316_cast_fp16 = reshape(shape = concat_1055, x = transpose_211_cast_fp16)[name = string("reshape_316_cast_fp16")]; bool matmul_105_transpose_x_0 = const()[name = string("matmul_105_transpose_x_0"), val = bool(false)]; bool matmul_105_transpose_y_0 = const()[name = string("matmul_105_transpose_y_0"), val = bool(false)]; tensor matmul_105_cast_fp16 = matmul(transpose_x = matmul_105_transpose_x_0, transpose_y = matmul_105_transpose_y_0, x = reshape_315_cast_fp16, y = reshape_316_cast_fp16)[name = string("matmul_105_cast_fp16")]; tensor concat_1059 = const()[name = string("concat_1059"), val = tensor([1, 1, 104, 104])]; tensor reshape_317_cast_fp16 = reshape(shape = concat_1059, x = matmul_105_cast_fp16)[name = string("reshape_317_cast_fp16")]; tensor transpose_2793_perm_0 = const()[name = string("transpose_2793_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2793 = transpose(perm = transpose_2793_perm_0, x = reshape_317_cast_fp16)[name = string("transpose_3906")]; tensor w_423_cast_fp16 = add(x = transpose_2793, y = transpose_2305)[name = string("w_423_cast_fp16")]; tensor var_2167_cast_fp16 = softmax(axis = var_1983, x = w_423_cast_fp16)[name = string("op_2167_cast_fp16")]; string var_2169_equation_0 = const()[name = string("op_2169_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2169_cast_fp16 = einsum(equation = var_2169_equation_0, values = (var_2073_cast_fp16_9, var_2167_cast_fp16))[name = string("op_2169_cast_fp16")]; tensor transpose_212_perm_0 = const()[name = string("transpose_212_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1064 = const()[name = string("concat_1064"), val = tensor([1, 104, 64])]; tensor transpose_212_cast_fp16 = transpose(perm = transpose_212_perm_0, x = var_2039_cast_fp16_10)[name = string("transpose_3905")]; tensor reshape_318_cast_fp16 = reshape(shape = concat_1064, x = transpose_212_cast_fp16)[name = string("reshape_318_cast_fp16")]; tensor transpose_213_perm_0 = const()[name = string("transpose_213_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1065 = const()[name = string("concat_1065"), val = tensor([1, 64, 104])]; tensor transpose_213_cast_fp16 = transpose(perm = transpose_213_perm_0, x = var_2056_cast_fp16_10)[name = string("transpose_3904")]; tensor reshape_319_cast_fp16 = reshape(shape = concat_1065, x = transpose_213_cast_fp16)[name = string("reshape_319_cast_fp16")]; bool matmul_106_transpose_x_0 = const()[name = string("matmul_106_transpose_x_0"), val = bool(false)]; bool matmul_106_transpose_y_0 = const()[name = string("matmul_106_transpose_y_0"), val = bool(false)]; tensor matmul_106_cast_fp16 = matmul(transpose_x = matmul_106_transpose_x_0, transpose_y = matmul_106_transpose_y_0, x = reshape_318_cast_fp16, y = reshape_319_cast_fp16)[name = string("matmul_106_cast_fp16")]; tensor concat_1069 = const()[name = string("concat_1069"), val = tensor([1, 1, 104, 104])]; tensor reshape_320_cast_fp16 = reshape(shape = concat_1069, x = matmul_106_cast_fp16)[name = string("reshape_320_cast_fp16")]; tensor transpose_2794_perm_0 = const()[name = string("transpose_2794_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2794 = transpose(perm = transpose_2794_perm_0, x = reshape_320_cast_fp16)[name = string("transpose_3903")]; tensor w_427_cast_fp16 = add(x = transpose_2794, y = transpose_2305)[name = string("w_427_cast_fp16")]; tensor var_2175_cast_fp16 = softmax(axis = var_1983, x = w_427_cast_fp16)[name = string("op_2175_cast_fp16")]; string var_2177_equation_0 = const()[name = string("op_2177_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2177_cast_fp16 = einsum(equation = var_2177_equation_0, values = (var_2073_cast_fp16_10, var_2175_cast_fp16))[name = string("op_2177_cast_fp16")]; tensor transpose_214_perm_0 = const()[name = string("transpose_214_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1074 = const()[name = string("concat_1074"), val = tensor([1, 104, 64])]; tensor transpose_214_cast_fp16 = transpose(perm = transpose_214_perm_0, x = var_2039_cast_fp16_11)[name = string("transpose_3902")]; tensor reshape_321_cast_fp16 = reshape(shape = concat_1074, x = transpose_214_cast_fp16)[name = string("reshape_321_cast_fp16")]; tensor transpose_215_perm_0 = const()[name = string("transpose_215_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1075 = const()[name = string("concat_1075"), val = tensor([1, 64, 104])]; tensor transpose_215_cast_fp16 = transpose(perm = transpose_215_perm_0, x = var_2056_cast_fp16_11)[name = string("transpose_3901")]; tensor reshape_322_cast_fp16 = reshape(shape = concat_1075, x = transpose_215_cast_fp16)[name = string("reshape_322_cast_fp16")]; bool matmul_107_transpose_x_0 = const()[name = string("matmul_107_transpose_x_0"), val = bool(false)]; bool matmul_107_transpose_y_0 = const()[name = string("matmul_107_transpose_y_0"), val = bool(false)]; tensor matmul_107_cast_fp16 = matmul(transpose_x = matmul_107_transpose_x_0, transpose_y = matmul_107_transpose_y_0, x = reshape_321_cast_fp16, y = reshape_322_cast_fp16)[name = string("matmul_107_cast_fp16")]; tensor concat_1079 = const()[name = string("concat_1079"), val = tensor([1, 1, 104, 104])]; tensor reshape_323_cast_fp16 = reshape(shape = concat_1079, x = matmul_107_cast_fp16)[name = string("reshape_323_cast_fp16")]; tensor transpose_2795_perm_0 = const()[name = string("transpose_2795_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2795 = transpose(perm = transpose_2795_perm_0, x = reshape_323_cast_fp16)[name = string("transpose_3900")]; tensor w_431_cast_fp16 = add(x = transpose_2795, y = transpose_2305)[name = string("w_431_cast_fp16")]; tensor var_2183_cast_fp16 = softmax(axis = var_1983, x = w_431_cast_fp16)[name = string("op_2183_cast_fp16")]; string var_2185_equation_0 = const()[name = string("op_2185_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2185_cast_fp16 = einsum(equation = var_2185_equation_0, values = (var_2073_cast_fp16_11, var_2183_cast_fp16))[name = string("op_2185_cast_fp16")]; tensor transpose_216_perm_0 = const()[name = string("transpose_216_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1084 = const()[name = string("concat_1084"), val = tensor([1, 104, 64])]; tensor transpose_216_cast_fp16 = transpose(perm = transpose_216_perm_0, x = var_2039_cast_fp16_12)[name = string("transpose_3899")]; tensor reshape_324_cast_fp16 = reshape(shape = concat_1084, x = transpose_216_cast_fp16)[name = string("reshape_324_cast_fp16")]; tensor transpose_217_perm_0 = const()[name = string("transpose_217_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1085 = const()[name = string("concat_1085"), val = tensor([1, 64, 104])]; tensor transpose_217_cast_fp16 = transpose(perm = transpose_217_perm_0, x = var_2056_cast_fp16_12)[name = string("transpose_3898")]; tensor reshape_325_cast_fp16 = reshape(shape = concat_1085, x = transpose_217_cast_fp16)[name = string("reshape_325_cast_fp16")]; bool matmul_108_transpose_x_0 = const()[name = string("matmul_108_transpose_x_0"), val = bool(false)]; bool matmul_108_transpose_y_0 = const()[name = string("matmul_108_transpose_y_0"), val = bool(false)]; tensor matmul_108_cast_fp16 = matmul(transpose_x = matmul_108_transpose_x_0, transpose_y = matmul_108_transpose_y_0, x = reshape_324_cast_fp16, y = reshape_325_cast_fp16)[name = string("matmul_108_cast_fp16")]; tensor concat_1089 = const()[name = string("concat_1089"), val = tensor([1, 1, 104, 104])]; tensor reshape_326_cast_fp16 = reshape(shape = concat_1089, x = matmul_108_cast_fp16)[name = string("reshape_326_cast_fp16")]; tensor transpose_2796_perm_0 = const()[name = string("transpose_2796_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2796 = transpose(perm = transpose_2796_perm_0, x = reshape_326_cast_fp16)[name = string("transpose_3897")]; tensor w_435_cast_fp16 = add(x = transpose_2796, y = transpose_2305)[name = string("w_435_cast_fp16")]; tensor var_2191_cast_fp16 = softmax(axis = var_1983, x = w_435_cast_fp16)[name = string("op_2191_cast_fp16")]; string var_2193_equation_0 = const()[name = string("op_2193_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2193_cast_fp16 = einsum(equation = var_2193_equation_0, values = (var_2073_cast_fp16_12, var_2191_cast_fp16))[name = string("op_2193_cast_fp16")]; tensor transpose_218_perm_0 = const()[name = string("transpose_218_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1094 = const()[name = string("concat_1094"), val = tensor([1, 104, 64])]; tensor transpose_218_cast_fp16 = transpose(perm = transpose_218_perm_0, x = var_2039_cast_fp16_13)[name = string("transpose_3896")]; tensor reshape_327_cast_fp16 = reshape(shape = concat_1094, x = transpose_218_cast_fp16)[name = string("reshape_327_cast_fp16")]; tensor transpose_219_perm_0 = const()[name = string("transpose_219_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1095 = const()[name = string("concat_1095"), val = tensor([1, 64, 104])]; tensor transpose_219_cast_fp16 = transpose(perm = transpose_219_perm_0, x = var_2056_cast_fp16_13)[name = string("transpose_3895")]; tensor reshape_328_cast_fp16 = reshape(shape = concat_1095, x = transpose_219_cast_fp16)[name = string("reshape_328_cast_fp16")]; bool matmul_109_transpose_x_0 = const()[name = string("matmul_109_transpose_x_0"), val = bool(false)]; bool matmul_109_transpose_y_0 = const()[name = string("matmul_109_transpose_y_0"), val = bool(false)]; tensor matmul_109_cast_fp16 = matmul(transpose_x = matmul_109_transpose_x_0, transpose_y = matmul_109_transpose_y_0, x = reshape_327_cast_fp16, y = reshape_328_cast_fp16)[name = string("matmul_109_cast_fp16")]; tensor concat_1099 = const()[name = string("concat_1099"), val = tensor([1, 1, 104, 104])]; tensor reshape_329_cast_fp16 = reshape(shape = concat_1099, x = matmul_109_cast_fp16)[name = string("reshape_329_cast_fp16")]; tensor transpose_2797_perm_0 = const()[name = string("transpose_2797_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2797 = transpose(perm = transpose_2797_perm_0, x = reshape_329_cast_fp16)[name = string("transpose_3894")]; tensor w_439_cast_fp16 = add(x = transpose_2797, y = transpose_2305)[name = string("w_439_cast_fp16")]; tensor var_2199_cast_fp16 = softmax(axis = var_1983, x = w_439_cast_fp16)[name = string("op_2199_cast_fp16")]; string var_2201_equation_0 = const()[name = string("op_2201_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2201_cast_fp16 = einsum(equation = var_2201_equation_0, values = (var_2073_cast_fp16_13, var_2199_cast_fp16))[name = string("op_2201_cast_fp16")]; tensor transpose_220_perm_0 = const()[name = string("transpose_220_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1104 = const()[name = string("concat_1104"), val = tensor([1, 104, 64])]; tensor transpose_220_cast_fp16 = transpose(perm = transpose_220_perm_0, x = var_2039_cast_fp16_14)[name = string("transpose_3893")]; tensor reshape_330_cast_fp16 = reshape(shape = concat_1104, x = transpose_220_cast_fp16)[name = string("reshape_330_cast_fp16")]; tensor transpose_221_perm_0 = const()[name = string("transpose_221_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1105 = const()[name = string("concat_1105"), val = tensor([1, 64, 104])]; tensor transpose_221_cast_fp16 = transpose(perm = transpose_221_perm_0, x = var_2056_cast_fp16_14)[name = string("transpose_3892")]; tensor reshape_331_cast_fp16 = reshape(shape = concat_1105, x = transpose_221_cast_fp16)[name = string("reshape_331_cast_fp16")]; bool matmul_110_transpose_x_0 = const()[name = string("matmul_110_transpose_x_0"), val = bool(false)]; bool matmul_110_transpose_y_0 = const()[name = string("matmul_110_transpose_y_0"), val = bool(false)]; tensor matmul_110_cast_fp16 = matmul(transpose_x = matmul_110_transpose_x_0, transpose_y = matmul_110_transpose_y_0, x = reshape_330_cast_fp16, y = reshape_331_cast_fp16)[name = string("matmul_110_cast_fp16")]; tensor concat_1109 = const()[name = string("concat_1109"), val = tensor([1, 1, 104, 104])]; tensor reshape_332_cast_fp16 = reshape(shape = concat_1109, x = matmul_110_cast_fp16)[name = string("reshape_332_cast_fp16")]; tensor transpose_2798_perm_0 = const()[name = string("transpose_2798_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2798 = transpose(perm = transpose_2798_perm_0, x = reshape_332_cast_fp16)[name = string("transpose_3891")]; tensor w_443_cast_fp16 = add(x = transpose_2798, y = transpose_2305)[name = string("w_443_cast_fp16")]; tensor var_2207_cast_fp16 = softmax(axis = var_1983, x = w_443_cast_fp16)[name = string("op_2207_cast_fp16")]; string var_2209_equation_0 = const()[name = string("op_2209_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2209_cast_fp16 = einsum(equation = var_2209_equation_0, values = (var_2073_cast_fp16_14, var_2207_cast_fp16))[name = string("op_2209_cast_fp16")]; tensor transpose_222_perm_0 = const()[name = string("transpose_222_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1114 = const()[name = string("concat_1114"), val = tensor([1, 104, 64])]; tensor transpose_222_cast_fp16 = transpose(perm = transpose_222_perm_0, x = var_2039_cast_fp16_15)[name = string("transpose_3890")]; tensor reshape_333_cast_fp16 = reshape(shape = concat_1114, x = transpose_222_cast_fp16)[name = string("reshape_333_cast_fp16")]; tensor transpose_223_perm_0 = const()[name = string("transpose_223_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1115 = const()[name = string("concat_1115"), val = tensor([1, 64, 104])]; tensor transpose_223_cast_fp16 = transpose(perm = transpose_223_perm_0, x = var_2056_cast_fp16_15)[name = string("transpose_3889")]; tensor reshape_334_cast_fp16 = reshape(shape = concat_1115, x = transpose_223_cast_fp16)[name = string("reshape_334_cast_fp16")]; bool matmul_111_transpose_x_0 = const()[name = string("matmul_111_transpose_x_0"), val = bool(false)]; bool matmul_111_transpose_y_0 = const()[name = string("matmul_111_transpose_y_0"), val = bool(false)]; tensor matmul_111_cast_fp16 = matmul(transpose_x = matmul_111_transpose_x_0, transpose_y = matmul_111_transpose_y_0, x = reshape_333_cast_fp16, y = reshape_334_cast_fp16)[name = string("matmul_111_cast_fp16")]; tensor concat_1119 = const()[name = string("concat_1119"), val = tensor([1, 1, 104, 104])]; tensor reshape_335_cast_fp16 = reshape(shape = concat_1119, x = matmul_111_cast_fp16)[name = string("reshape_335_cast_fp16")]; tensor transpose_2799_perm_0 = const()[name = string("transpose_2799_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2799 = transpose(perm = transpose_2799_perm_0, x = reshape_335_cast_fp16)[name = string("transpose_3888")]; tensor w_447_cast_fp16 = add(x = transpose_2799, y = transpose_2305)[name = string("w_447_cast_fp16")]; tensor var_2215_cast_fp16 = softmax(axis = var_1983, x = w_447_cast_fp16)[name = string("op_2215_cast_fp16")]; string var_2217_equation_0 = const()[name = string("op_2217_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2217_cast_fp16 = einsum(equation = var_2217_equation_0, values = (var_2073_cast_fp16_15, var_2215_cast_fp16))[name = string("op_2217_cast_fp16")]; bool input_59_interleave_0 = const()[name = string("input_59_interleave_0"), val = bool(false)]; tensor input_59_cast_fp16 = concat(axis = var_1983, interleave = input_59_interleave_0, values = (var_2097_cast_fp16, var_2105_cast_fp16, var_2113_cast_fp16, var_2121_cast_fp16, var_2129_cast_fp16, var_2137_cast_fp16, var_2145_cast_fp16, var_2153_cast_fp16, var_2161_cast_fp16, var_2169_cast_fp16, var_2177_cast_fp16, var_2185_cast_fp16, var_2193_cast_fp16, var_2201_cast_fp16, var_2209_cast_fp16, var_2217_cast_fp16))[name = string("input_59_cast_fp16")]; string var_2226_pad_type_0 = const()[name = string("op_2226_pad_type_0"), val = string("valid")]; tensor var_2226_strides_0 = const()[name = string("op_2226_strides_0"), val = tensor([1, 1])]; tensor var_2226_pad_0 = const()[name = string("op_2226_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2226_dilations_0 = const()[name = string("op_2226_dilations_0"), val = tensor([1, 1])]; int32 var_2226_groups_0 = const()[name = string("op_2226_groups_0"), val = int32(1)]; tensor layers_6_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_6_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(181531712)))]; tensor layers_6_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_6_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(183628928)))]; tensor var_2226_cast_fp16 = conv(bias = layers_6_self_attn_out_proj_bias_to_fp16, dilations = var_2226_dilations_0, groups = var_2226_groups_0, pad = var_2226_pad_0, pad_type = var_2226_pad_type_0, strides = var_2226_strides_0, weight = layers_6_self_attn_out_proj_weight_to_fp16, x = input_59_cast_fp16)[name = string("op_2226_cast_fp16")]; tensor x_77_cast_fp16 = add(x = x_73_cast_fp16, y = var_2226_cast_fp16)[name = string("x_77_cast_fp16")]; tensor mu_27_axes_0 = const()[name = string("mu_27_axes_0"), val = tensor([1])]; bool mu_27_keep_dims_0 = const()[name = string("mu_27_keep_dims_0"), val = bool(true)]; tensor mu_27_cast_fp16 = reduce_mean(axes = mu_27_axes_0, keep_dims = mu_27_keep_dims_0, x = x_77_cast_fp16)[name = string("mu_27_cast_fp16")]; tensor var_2232_cast_fp16 = sub(x = x_77_cast_fp16, y = mu_27_cast_fp16)[name = string("op_2232_cast_fp16")]; fp16 var_1986_promoted_1_to_fp16 = const()[name = string("op_1986_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_2233_cast_fp16 = pow(x = var_2232_cast_fp16, y = var_1986_promoted_1_to_fp16)[name = string("op_2233_cast_fp16")]; tensor var_27_axes_0 = const()[name = string("var_27_axes_0"), val = tensor([1])]; bool var_27_keep_dims_0 = const()[name = string("var_27_keep_dims_0"), val = bool(true)]; tensor var_27_cast_fp16 = reduce_mean(axes = var_27_axes_0, keep_dims = var_27_keep_dims_0, x = var_2233_cast_fp16)[name = string("var_27_cast_fp16")]; fp16 var_2237_to_fp16 = const()[name = string("op_2237_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2238_cast_fp16 = add(x = var_27_cast_fp16, y = var_2237_to_fp16)[name = string("op_2238_cast_fp16")]; fp32 var_2239_epsilon_0 = const()[name = string("op_2239_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2239_cast_fp16 = rsqrt(epsilon = var_2239_epsilon_0, x = var_2238_cast_fp16)[name = string("op_2239_cast_fp16")]; tensor x_79_cast_fp16 = mul(x = var_2232_cast_fp16, y = var_2239_cast_fp16)[name = string("x_79_cast_fp16")]; tensor input_61_gamma_0_to_fp16 = const()[name = string("input_61_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(183631040)))]; tensor input_61_beta_0_to_fp16 = const()[name = string("input_61_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(183633152)))]; fp16 input_61_epsilon_0_to_fp16 = const()[name = string("input_61_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_61_cast_fp16 = batch_norm(beta = input_61_beta_0_to_fp16, epsilon = input_61_epsilon_0_to_fp16, gamma = input_61_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_79_cast_fp16)[name = string("input_61_cast_fp16")]; string x_81_pad_type_0 = const()[name = string("x_81_pad_type_0"), val = string("valid")]; tensor x_81_strides_0 = const()[name = string("x_81_strides_0"), val = tensor([1, 1])]; tensor x_81_pad_0 = const()[name = string("x_81_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_81_dilations_0 = const()[name = string("x_81_dilations_0"), val = tensor([1, 1])]; int32 x_81_groups_0 = const()[name = string("x_81_groups_0"), val = int32(1)]; tensor layers_6_fc1_weight_to_fp16 = const()[name = string("layers_6_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(183635264)))]; tensor layers_6_fc1_bias_to_fp16 = const()[name = string("layers_6_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(192023936)))]; tensor x_81_cast_fp16 = conv(bias = layers_6_fc1_bias_to_fp16, dilations = x_81_dilations_0, groups = x_81_groups_0, pad = x_81_pad_0, pad_type = x_81_pad_type_0, strides = x_81_strides_0, weight = layers_6_fc1_weight_to_fp16, x = input_61_cast_fp16)[name = string("x_81_cast_fp16")]; fp16 var_2254_to_fp16 = const()[name = string("op_2254_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_2255_cast_fp16 = mul(x = x_81_cast_fp16, y = var_2254_to_fp16)[name = string("op_2255_cast_fp16")]; tensor var_2256_cast_fp16 = mul(x = var_2255_cast_fp16, y = x_81_cast_fp16)[name = string("op_2256_cast_fp16")]; tensor var_2257_cast_fp16 = mul(x = var_2256_cast_fp16, y = x_81_cast_fp16)[name = string("op_2257_cast_fp16")]; tensor var_2258_cast_fp16 = add(x = x_81_cast_fp16, y = var_2257_cast_fp16)[name = string("op_2258_cast_fp16")]; fp16 var_2259_to_fp16 = const()[name = string("op_2259_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_19_cast_fp16 = mul(x = var_2258_cast_fp16, y = var_2259_to_fp16)[name = string("u_19_cast_fp16")]; fp16 var_2261_to_fp16 = const()[name = string("op_2261_to_fp16"), val = fp16(0x1p-1)]; tensor var_2262_cast_fp16 = mul(x = x_81_cast_fp16, y = var_2261_to_fp16)[name = string("op_2262_cast_fp16")]; tensor var_2263_cast_fp16 = tanh(x = u_19_cast_fp16)[name = string("op_2263_cast_fp16")]; fp16 var_2264_to_fp16 = const()[name = string("op_2264_to_fp16"), val = fp16(0x1p+0)]; tensor var_2265_cast_fp16 = add(x = var_2263_cast_fp16, y = var_2264_to_fp16)[name = string("op_2265_cast_fp16")]; tensor input_63_cast_fp16 = mul(x = var_2262_cast_fp16, y = var_2265_cast_fp16)[name = string("input_63_cast_fp16")]; string h_13_pad_type_0 = const()[name = string("h_13_pad_type_0"), val = string("valid")]; tensor h_13_strides_0 = const()[name = string("h_13_strides_0"), val = tensor([1, 1])]; tensor h_13_pad_0 = const()[name = string("h_13_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_13_dilations_0 = const()[name = string("h_13_dilations_0"), val = tensor([1, 1])]; int32 h_13_groups_0 = const()[name = string("h_13_groups_0"), val = int32(1)]; tensor layers_6_fc2_weight_to_fp16 = const()[name = string("layers_6_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(192032192)))]; tensor layers_6_fc2_bias_to_fp16 = const()[name = string("layers_6_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(200420864)))]; tensor h_13_cast_fp16 = conv(bias = layers_6_fc2_bias_to_fp16, dilations = h_13_dilations_0, groups = h_13_groups_0, pad = h_13_pad_0, pad_type = h_13_pad_type_0, strides = h_13_strides_0, weight = layers_6_fc2_weight_to_fp16, x = input_63_cast_fp16)[name = string("h_13_cast_fp16")]; tensor x_83_cast_fp16 = add(x = x_77_cast_fp16, y = h_13_cast_fp16)[name = string("x_83_cast_fp16")]; int32 var_2281 = const()[name = string("op_2281"), val = int32(1)]; tensor mu_29_axes_0 = const()[name = string("mu_29_axes_0"), val = tensor([1])]; bool mu_29_keep_dims_0 = const()[name = string("mu_29_keep_dims_0"), val = bool(true)]; tensor mu_29_cast_fp16 = reduce_mean(axes = mu_29_axes_0, keep_dims = mu_29_keep_dims_0, x = x_83_cast_fp16)[name = string("mu_29_cast_fp16")]; tensor var_2295_cast_fp16 = sub(x = x_83_cast_fp16, y = mu_29_cast_fp16)[name = string("op_2295_cast_fp16")]; fp16 var_2284_promoted_to_fp16 = const()[name = string("op_2284_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_2296_cast_fp16 = pow(x = var_2295_cast_fp16, y = var_2284_promoted_to_fp16)[name = string("op_2296_cast_fp16")]; tensor var_29_axes_0 = const()[name = string("var_29_axes_0"), val = tensor([1])]; bool var_29_keep_dims_0 = const()[name = string("var_29_keep_dims_0"), val = bool(true)]; tensor var_29_cast_fp16 = reduce_mean(axes = var_29_axes_0, keep_dims = var_29_keep_dims_0, x = var_2296_cast_fp16)[name = string("var_29_cast_fp16")]; fp16 var_2300_to_fp16 = const()[name = string("op_2300_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2301_cast_fp16 = add(x = var_29_cast_fp16, y = var_2300_to_fp16)[name = string("op_2301_cast_fp16")]; fp32 var_2302_epsilon_0 = const()[name = string("op_2302_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2302_cast_fp16 = rsqrt(epsilon = var_2302_epsilon_0, x = var_2301_cast_fp16)[name = string("op_2302_cast_fp16")]; tensor x_85_cast_fp16 = mul(x = var_2295_cast_fp16, y = var_2302_cast_fp16)[name = string("x_85_cast_fp16")]; tensor input_65_gamma_0_to_fp16 = const()[name = string("input_65_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(200422976)))]; tensor input_65_beta_0_to_fp16 = const()[name = string("input_65_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(200425088)))]; fp16 input_65_epsilon_0_to_fp16 = const()[name = string("input_65_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_65_cast_fp16 = batch_norm(beta = input_65_beta_0_to_fp16, epsilon = input_65_epsilon_0_to_fp16, gamma = input_65_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_85_cast_fp16)[name = string("input_65_cast_fp16")]; string var_2320_pad_type_0 = const()[name = string("op_2320_pad_type_0"), val = string("valid")]; tensor var_2320_strides_0 = const()[name = string("op_2320_strides_0"), val = tensor([1, 1])]; tensor var_2320_pad_0 = const()[name = string("op_2320_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2320_dilations_0 = const()[name = string("op_2320_dilations_0"), val = tensor([1, 1])]; int32 var_2320_groups_0 = const()[name = string("op_2320_groups_0"), val = int32(1)]; tensor var_2322_weight_0_to_fp16 = const()[name = string("op_2322_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(200427200)))]; tensor var_2322_bias_0_to_fp16 = const()[name = string("op_2322_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(202524416)))]; tensor var_2322_cast_fp16 = conv(bias = var_2322_bias_0_to_fp16, dilations = var_2320_dilations_0, groups = var_2320_groups_0, pad = var_2320_pad_0, pad_type = var_2320_pad_type_0, strides = var_2320_strides_0, weight = var_2322_weight_0_to_fp16, x = input_65_cast_fp16)[name = string("op_2322_cast_fp16")]; string var_2329_pad_type_0 = const()[name = string("op_2329_pad_type_0"), val = string("valid")]; tensor var_2329_strides_0 = const()[name = string("op_2329_strides_0"), val = tensor([1, 1])]; tensor var_2329_pad_0 = const()[name = string("op_2329_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2329_dilations_0 = const()[name = string("op_2329_dilations_0"), val = tensor([1, 1])]; int32 var_2329_groups_0 = const()[name = string("op_2329_groups_0"), val = int32(1)]; tensor layers_7_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_7_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(202526528)))]; tensor layers_7_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_7_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(204623744)))]; tensor var_2329_cast_fp16 = conv(bias = layers_7_self_attn_k_proj_bias_to_fp16, dilations = var_2329_dilations_0, groups = var_2329_groups_0, pad = var_2329_pad_0, pad_type = var_2329_pad_type_0, strides = var_2329_strides_0, weight = layers_7_self_attn_k_proj_weight_to_fp16, x = input_65_cast_fp16)[name = string("op_2329_cast_fp16")]; string var_2336_pad_type_0 = const()[name = string("op_2336_pad_type_0"), val = string("valid")]; tensor var_2336_strides_0 = const()[name = string("op_2336_strides_0"), val = tensor([1, 1])]; tensor var_2336_pad_0 = const()[name = string("op_2336_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2336_dilations_0 = const()[name = string("op_2336_dilations_0"), val = tensor([1, 1])]; int32 var_2336_groups_0 = const()[name = string("op_2336_groups_0"), val = int32(1)]; tensor layers_7_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_7_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(204625856)))]; tensor layers_7_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_7_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(206723072)))]; tensor var_2336_cast_fp16 = conv(bias = layers_7_self_attn_v_proj_bias_to_fp16, dilations = var_2336_dilations_0, groups = var_2336_groups_0, pad = var_2336_pad_0, pad_type = var_2336_pad_type_0, strides = var_2336_strides_0, weight = layers_7_self_attn_v_proj_weight_to_fp16, x = input_65_cast_fp16)[name = string("op_2336_cast_fp16")]; tensor tile_21 = const()[name = string("tile_21"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(206725184)))]; int32 var_2337_axis_0 = const()[name = string("op_2337_axis_0"), val = int32(1)]; tensor var_2337_cast_fp16_0, tensor var_2337_cast_fp16_1, tensor var_2337_cast_fp16_2, tensor var_2337_cast_fp16_3, tensor var_2337_cast_fp16_4, tensor var_2337_cast_fp16_5, tensor var_2337_cast_fp16_6, tensor var_2337_cast_fp16_7, tensor var_2337_cast_fp16_8, tensor var_2337_cast_fp16_9, tensor var_2337_cast_fp16_10, tensor var_2337_cast_fp16_11, tensor var_2337_cast_fp16_12, tensor var_2337_cast_fp16_13, tensor var_2337_cast_fp16_14, tensor var_2337_cast_fp16_15 = split(axis = var_2337_axis_0, split_sizes = tile_21, x = var_2322_cast_fp16)[name = string("op_2337_cast_fp16")]; tensor tile_22 = const()[name = string("tile_22"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(206725312)))]; int32 var_2354_axis_0 = const()[name = string("op_2354_axis_0"), val = int32(1)]; tensor var_2354_cast_fp16_0, tensor var_2354_cast_fp16_1, tensor var_2354_cast_fp16_2, tensor var_2354_cast_fp16_3, tensor var_2354_cast_fp16_4, tensor var_2354_cast_fp16_5, tensor var_2354_cast_fp16_6, tensor var_2354_cast_fp16_7, tensor var_2354_cast_fp16_8, tensor var_2354_cast_fp16_9, tensor var_2354_cast_fp16_10, tensor var_2354_cast_fp16_11, tensor var_2354_cast_fp16_12, tensor var_2354_cast_fp16_13, tensor var_2354_cast_fp16_14, tensor var_2354_cast_fp16_15 = split(axis = var_2354_axis_0, split_sizes = tile_22, x = var_2329_cast_fp16)[name = string("op_2354_cast_fp16")]; tensor tile_23 = const()[name = string("tile_23"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(206725440)))]; int32 var_2371_axis_0 = const()[name = string("op_2371_axis_0"), val = int32(1)]; tensor var_2371_cast_fp16_0, tensor var_2371_cast_fp16_1, tensor var_2371_cast_fp16_2, tensor var_2371_cast_fp16_3, tensor var_2371_cast_fp16_4, tensor var_2371_cast_fp16_5, tensor var_2371_cast_fp16_6, tensor var_2371_cast_fp16_7, tensor var_2371_cast_fp16_8, tensor var_2371_cast_fp16_9, tensor var_2371_cast_fp16_10, tensor var_2371_cast_fp16_11, tensor var_2371_cast_fp16_12, tensor var_2371_cast_fp16_13, tensor var_2371_cast_fp16_14, tensor var_2371_cast_fp16_15 = split(axis = var_2371_axis_0, split_sizes = tile_23, x = var_2336_cast_fp16)[name = string("op_2371_cast_fp16")]; tensor transpose_224_perm_0 = const()[name = string("transpose_224_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1124 = const()[name = string("concat_1124"), val = tensor([1, 104, 64])]; tensor transpose_224_cast_fp16 = transpose(perm = transpose_224_perm_0, x = var_2337_cast_fp16_0)[name = string("transpose_3887")]; tensor reshape_336_cast_fp16 = reshape(shape = concat_1124, x = transpose_224_cast_fp16)[name = string("reshape_336_cast_fp16")]; tensor transpose_225_perm_0 = const()[name = string("transpose_225_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1125 = const()[name = string("concat_1125"), val = tensor([1, 64, 104])]; tensor transpose_225_cast_fp16 = transpose(perm = transpose_225_perm_0, x = var_2354_cast_fp16_0)[name = string("transpose_3886")]; tensor reshape_337_cast_fp16 = reshape(shape = concat_1125, x = transpose_225_cast_fp16)[name = string("reshape_337_cast_fp16")]; bool matmul_112_transpose_x_0 = const()[name = string("matmul_112_transpose_x_0"), val = bool(false)]; bool matmul_112_transpose_y_0 = const()[name = string("matmul_112_transpose_y_0"), val = bool(false)]; tensor matmul_112_cast_fp16 = matmul(transpose_x = matmul_112_transpose_x_0, transpose_y = matmul_112_transpose_y_0, x = reshape_336_cast_fp16, y = reshape_337_cast_fp16)[name = string("matmul_112_cast_fp16")]; tensor concat_1129 = const()[name = string("concat_1129"), val = tensor([1, 1, 104, 104])]; tensor reshape_338_cast_fp16 = reshape(shape = concat_1129, x = matmul_112_cast_fp16)[name = string("reshape_338_cast_fp16")]; tensor transpose_2800_perm_0 = const()[name = string("transpose_2800_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2800 = transpose(perm = transpose_2800_perm_0, x = reshape_338_cast_fp16)[name = string("transpose_3885")]; tensor w_451_cast_fp16 = add(x = transpose_2800, y = transpose_2305)[name = string("w_451_cast_fp16")]; tensor var_2393_cast_fp16 = softmax(axis = var_2281, x = w_451_cast_fp16)[name = string("op_2393_cast_fp16")]; string var_2395_equation_0 = const()[name = string("op_2395_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2395_cast_fp16 = einsum(equation = var_2395_equation_0, values = (var_2371_cast_fp16_0, var_2393_cast_fp16))[name = string("op_2395_cast_fp16")]; tensor transpose_226_perm_0 = const()[name = string("transpose_226_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1134 = const()[name = string("concat_1134"), val = tensor([1, 104, 64])]; tensor transpose_226_cast_fp16 = transpose(perm = transpose_226_perm_0, x = var_2337_cast_fp16_1)[name = string("transpose_3884")]; tensor reshape_339_cast_fp16 = reshape(shape = concat_1134, x = transpose_226_cast_fp16)[name = string("reshape_339_cast_fp16")]; tensor transpose_227_perm_0 = const()[name = string("transpose_227_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1135 = const()[name = string("concat_1135"), val = tensor([1, 64, 104])]; tensor transpose_227_cast_fp16 = transpose(perm = transpose_227_perm_0, x = var_2354_cast_fp16_1)[name = string("transpose_3883")]; tensor reshape_340_cast_fp16 = reshape(shape = concat_1135, x = transpose_227_cast_fp16)[name = string("reshape_340_cast_fp16")]; bool matmul_113_transpose_x_0 = const()[name = string("matmul_113_transpose_x_0"), val = bool(false)]; bool matmul_113_transpose_y_0 = const()[name = string("matmul_113_transpose_y_0"), val = bool(false)]; tensor matmul_113_cast_fp16 = matmul(transpose_x = matmul_113_transpose_x_0, transpose_y = matmul_113_transpose_y_0, x = reshape_339_cast_fp16, y = reshape_340_cast_fp16)[name = string("matmul_113_cast_fp16")]; tensor concat_1139 = const()[name = string("concat_1139"), val = tensor([1, 1, 104, 104])]; tensor reshape_341_cast_fp16 = reshape(shape = concat_1139, x = matmul_113_cast_fp16)[name = string("reshape_341_cast_fp16")]; tensor transpose_2801_perm_0 = const()[name = string("transpose_2801_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2801 = transpose(perm = transpose_2801_perm_0, x = reshape_341_cast_fp16)[name = string("transpose_3882")]; tensor w_455_cast_fp16 = add(x = transpose_2801, y = transpose_2305)[name = string("w_455_cast_fp16")]; tensor var_2401_cast_fp16 = softmax(axis = var_2281, x = w_455_cast_fp16)[name = string("op_2401_cast_fp16")]; string var_2403_equation_0 = const()[name = string("op_2403_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2403_cast_fp16 = einsum(equation = var_2403_equation_0, values = (var_2371_cast_fp16_1, var_2401_cast_fp16))[name = string("op_2403_cast_fp16")]; tensor transpose_228_perm_0 = const()[name = string("transpose_228_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1144 = const()[name = string("concat_1144"), val = tensor([1, 104, 64])]; tensor transpose_228_cast_fp16 = transpose(perm = transpose_228_perm_0, x = var_2337_cast_fp16_2)[name = string("transpose_3881")]; tensor reshape_342_cast_fp16 = reshape(shape = concat_1144, x = transpose_228_cast_fp16)[name = string("reshape_342_cast_fp16")]; tensor transpose_229_perm_0 = const()[name = string("transpose_229_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1145 = const()[name = string("concat_1145"), val = tensor([1, 64, 104])]; tensor transpose_229_cast_fp16 = transpose(perm = transpose_229_perm_0, x = var_2354_cast_fp16_2)[name = string("transpose_3880")]; tensor reshape_343_cast_fp16 = reshape(shape = concat_1145, x = transpose_229_cast_fp16)[name = string("reshape_343_cast_fp16")]; bool matmul_114_transpose_x_0 = const()[name = string("matmul_114_transpose_x_0"), val = bool(false)]; bool matmul_114_transpose_y_0 = const()[name = string("matmul_114_transpose_y_0"), val = bool(false)]; tensor matmul_114_cast_fp16 = matmul(transpose_x = matmul_114_transpose_x_0, transpose_y = matmul_114_transpose_y_0, x = reshape_342_cast_fp16, y = reshape_343_cast_fp16)[name = string("matmul_114_cast_fp16")]; tensor concat_1149 = const()[name = string("concat_1149"), val = tensor([1, 1, 104, 104])]; tensor reshape_344_cast_fp16 = reshape(shape = concat_1149, x = matmul_114_cast_fp16)[name = string("reshape_344_cast_fp16")]; tensor transpose_2802_perm_0 = const()[name = string("transpose_2802_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2802 = transpose(perm = transpose_2802_perm_0, x = reshape_344_cast_fp16)[name = string("transpose_3879")]; tensor w_459_cast_fp16 = add(x = transpose_2802, y = transpose_2305)[name = string("w_459_cast_fp16")]; tensor var_2409_cast_fp16 = softmax(axis = var_2281, x = w_459_cast_fp16)[name = string("op_2409_cast_fp16")]; string var_2411_equation_0 = const()[name = string("op_2411_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2411_cast_fp16 = einsum(equation = var_2411_equation_0, values = (var_2371_cast_fp16_2, var_2409_cast_fp16))[name = string("op_2411_cast_fp16")]; tensor transpose_230_perm_0 = const()[name = string("transpose_230_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1154 = const()[name = string("concat_1154"), val = tensor([1, 104, 64])]; tensor transpose_230_cast_fp16 = transpose(perm = transpose_230_perm_0, x = var_2337_cast_fp16_3)[name = string("transpose_3878")]; tensor reshape_345_cast_fp16 = reshape(shape = concat_1154, x = transpose_230_cast_fp16)[name = string("reshape_345_cast_fp16")]; tensor transpose_231_perm_0 = const()[name = string("transpose_231_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1155 = const()[name = string("concat_1155"), val = tensor([1, 64, 104])]; tensor transpose_231_cast_fp16 = transpose(perm = transpose_231_perm_0, x = var_2354_cast_fp16_3)[name = string("transpose_3877")]; tensor reshape_346_cast_fp16 = reshape(shape = concat_1155, x = transpose_231_cast_fp16)[name = string("reshape_346_cast_fp16")]; bool matmul_115_transpose_x_0 = const()[name = string("matmul_115_transpose_x_0"), val = bool(false)]; bool matmul_115_transpose_y_0 = const()[name = string("matmul_115_transpose_y_0"), val = bool(false)]; tensor matmul_115_cast_fp16 = matmul(transpose_x = matmul_115_transpose_x_0, transpose_y = matmul_115_transpose_y_0, x = reshape_345_cast_fp16, y = reshape_346_cast_fp16)[name = string("matmul_115_cast_fp16")]; tensor concat_1159 = const()[name = string("concat_1159"), val = tensor([1, 1, 104, 104])]; tensor reshape_347_cast_fp16 = reshape(shape = concat_1159, x = matmul_115_cast_fp16)[name = string("reshape_347_cast_fp16")]; tensor transpose_2803_perm_0 = const()[name = string("transpose_2803_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2803 = transpose(perm = transpose_2803_perm_0, x = reshape_347_cast_fp16)[name = string("transpose_3876")]; tensor w_463_cast_fp16 = add(x = transpose_2803, y = transpose_2305)[name = string("w_463_cast_fp16")]; tensor var_2417_cast_fp16 = softmax(axis = var_2281, x = w_463_cast_fp16)[name = string("op_2417_cast_fp16")]; string var_2419_equation_0 = const()[name = string("op_2419_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2419_cast_fp16 = einsum(equation = var_2419_equation_0, values = (var_2371_cast_fp16_3, var_2417_cast_fp16))[name = string("op_2419_cast_fp16")]; tensor transpose_232_perm_0 = const()[name = string("transpose_232_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1164 = const()[name = string("concat_1164"), val = tensor([1, 104, 64])]; tensor transpose_232_cast_fp16 = transpose(perm = transpose_232_perm_0, x = var_2337_cast_fp16_4)[name = string("transpose_3875")]; tensor reshape_348_cast_fp16 = reshape(shape = concat_1164, x = transpose_232_cast_fp16)[name = string("reshape_348_cast_fp16")]; tensor transpose_233_perm_0 = const()[name = string("transpose_233_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1165 = const()[name = string("concat_1165"), val = tensor([1, 64, 104])]; tensor transpose_233_cast_fp16 = transpose(perm = transpose_233_perm_0, x = var_2354_cast_fp16_4)[name = string("transpose_3874")]; tensor reshape_349_cast_fp16 = reshape(shape = concat_1165, x = transpose_233_cast_fp16)[name = string("reshape_349_cast_fp16")]; bool matmul_116_transpose_x_0 = const()[name = string("matmul_116_transpose_x_0"), val = bool(false)]; bool matmul_116_transpose_y_0 = const()[name = string("matmul_116_transpose_y_0"), val = bool(false)]; tensor matmul_116_cast_fp16 = matmul(transpose_x = matmul_116_transpose_x_0, transpose_y = matmul_116_transpose_y_0, x = reshape_348_cast_fp16, y = reshape_349_cast_fp16)[name = string("matmul_116_cast_fp16")]; tensor concat_1169 = const()[name = string("concat_1169"), val = tensor([1, 1, 104, 104])]; tensor reshape_350_cast_fp16 = reshape(shape = concat_1169, x = matmul_116_cast_fp16)[name = string("reshape_350_cast_fp16")]; tensor transpose_2804_perm_0 = const()[name = string("transpose_2804_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2804 = transpose(perm = transpose_2804_perm_0, x = reshape_350_cast_fp16)[name = string("transpose_3873")]; tensor w_467_cast_fp16 = add(x = transpose_2804, y = transpose_2305)[name = string("w_467_cast_fp16")]; tensor var_2425_cast_fp16 = softmax(axis = var_2281, x = w_467_cast_fp16)[name = string("op_2425_cast_fp16")]; string var_2427_equation_0 = const()[name = string("op_2427_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2427_cast_fp16 = einsum(equation = var_2427_equation_0, values = (var_2371_cast_fp16_4, var_2425_cast_fp16))[name = string("op_2427_cast_fp16")]; tensor transpose_234_perm_0 = const()[name = string("transpose_234_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1174 = const()[name = string("concat_1174"), val = tensor([1, 104, 64])]; tensor transpose_234_cast_fp16 = transpose(perm = transpose_234_perm_0, x = var_2337_cast_fp16_5)[name = string("transpose_3872")]; tensor reshape_351_cast_fp16 = reshape(shape = concat_1174, x = transpose_234_cast_fp16)[name = string("reshape_351_cast_fp16")]; tensor transpose_235_perm_0 = const()[name = string("transpose_235_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1175 = const()[name = string("concat_1175"), val = tensor([1, 64, 104])]; tensor transpose_235_cast_fp16 = transpose(perm = transpose_235_perm_0, x = var_2354_cast_fp16_5)[name = string("transpose_3871")]; tensor reshape_352_cast_fp16 = reshape(shape = concat_1175, x = transpose_235_cast_fp16)[name = string("reshape_352_cast_fp16")]; bool matmul_117_transpose_x_0 = const()[name = string("matmul_117_transpose_x_0"), val = bool(false)]; bool matmul_117_transpose_y_0 = const()[name = string("matmul_117_transpose_y_0"), val = bool(false)]; tensor matmul_117_cast_fp16 = matmul(transpose_x = matmul_117_transpose_x_0, transpose_y = matmul_117_transpose_y_0, x = reshape_351_cast_fp16, y = reshape_352_cast_fp16)[name = string("matmul_117_cast_fp16")]; tensor concat_1179 = const()[name = string("concat_1179"), val = tensor([1, 1, 104, 104])]; tensor reshape_353_cast_fp16 = reshape(shape = concat_1179, x = matmul_117_cast_fp16)[name = string("reshape_353_cast_fp16")]; tensor transpose_2805_perm_0 = const()[name = string("transpose_2805_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2805 = transpose(perm = transpose_2805_perm_0, x = reshape_353_cast_fp16)[name = string("transpose_3870")]; tensor w_471_cast_fp16 = add(x = transpose_2805, y = transpose_2305)[name = string("w_471_cast_fp16")]; tensor var_2433_cast_fp16 = softmax(axis = var_2281, x = w_471_cast_fp16)[name = string("op_2433_cast_fp16")]; string var_2435_equation_0 = const()[name = string("op_2435_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2435_cast_fp16 = einsum(equation = var_2435_equation_0, values = (var_2371_cast_fp16_5, var_2433_cast_fp16))[name = string("op_2435_cast_fp16")]; tensor transpose_236_perm_0 = const()[name = string("transpose_236_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1184 = const()[name = string("concat_1184"), val = tensor([1, 104, 64])]; tensor transpose_236_cast_fp16 = transpose(perm = transpose_236_perm_0, x = var_2337_cast_fp16_6)[name = string("transpose_3869")]; tensor reshape_354_cast_fp16 = reshape(shape = concat_1184, x = transpose_236_cast_fp16)[name = string("reshape_354_cast_fp16")]; tensor transpose_237_perm_0 = const()[name = string("transpose_237_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1185 = const()[name = string("concat_1185"), val = tensor([1, 64, 104])]; tensor transpose_237_cast_fp16 = transpose(perm = transpose_237_perm_0, x = var_2354_cast_fp16_6)[name = string("transpose_3868")]; tensor reshape_355_cast_fp16 = reshape(shape = concat_1185, x = transpose_237_cast_fp16)[name = string("reshape_355_cast_fp16")]; bool matmul_118_transpose_x_0 = const()[name = string("matmul_118_transpose_x_0"), val = bool(false)]; bool matmul_118_transpose_y_0 = const()[name = string("matmul_118_transpose_y_0"), val = bool(false)]; tensor matmul_118_cast_fp16 = matmul(transpose_x = matmul_118_transpose_x_0, transpose_y = matmul_118_transpose_y_0, x = reshape_354_cast_fp16, y = reshape_355_cast_fp16)[name = string("matmul_118_cast_fp16")]; tensor concat_1189 = const()[name = string("concat_1189"), val = tensor([1, 1, 104, 104])]; tensor reshape_356_cast_fp16 = reshape(shape = concat_1189, x = matmul_118_cast_fp16)[name = string("reshape_356_cast_fp16")]; tensor transpose_2806_perm_0 = const()[name = string("transpose_2806_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2806 = transpose(perm = transpose_2806_perm_0, x = reshape_356_cast_fp16)[name = string("transpose_3867")]; tensor w_475_cast_fp16 = add(x = transpose_2806, y = transpose_2305)[name = string("w_475_cast_fp16")]; tensor var_2441_cast_fp16 = softmax(axis = var_2281, x = w_475_cast_fp16)[name = string("op_2441_cast_fp16")]; string var_2443_equation_0 = const()[name = string("op_2443_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2443_cast_fp16 = einsum(equation = var_2443_equation_0, values = (var_2371_cast_fp16_6, var_2441_cast_fp16))[name = string("op_2443_cast_fp16")]; tensor transpose_238_perm_0 = const()[name = string("transpose_238_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1194 = const()[name = string("concat_1194"), val = tensor([1, 104, 64])]; tensor transpose_238_cast_fp16 = transpose(perm = transpose_238_perm_0, x = var_2337_cast_fp16_7)[name = string("transpose_3866")]; tensor reshape_357_cast_fp16 = reshape(shape = concat_1194, x = transpose_238_cast_fp16)[name = string("reshape_357_cast_fp16")]; tensor transpose_239_perm_0 = const()[name = string("transpose_239_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1195 = const()[name = string("concat_1195"), val = tensor([1, 64, 104])]; tensor transpose_239_cast_fp16 = transpose(perm = transpose_239_perm_0, x = var_2354_cast_fp16_7)[name = string("transpose_3865")]; tensor reshape_358_cast_fp16 = reshape(shape = concat_1195, x = transpose_239_cast_fp16)[name = string("reshape_358_cast_fp16")]; bool matmul_119_transpose_x_0 = const()[name = string("matmul_119_transpose_x_0"), val = bool(false)]; bool matmul_119_transpose_y_0 = const()[name = string("matmul_119_transpose_y_0"), val = bool(false)]; tensor matmul_119_cast_fp16 = matmul(transpose_x = matmul_119_transpose_x_0, transpose_y = matmul_119_transpose_y_0, x = reshape_357_cast_fp16, y = reshape_358_cast_fp16)[name = string("matmul_119_cast_fp16")]; tensor concat_1199 = const()[name = string("concat_1199"), val = tensor([1, 1, 104, 104])]; tensor reshape_359_cast_fp16 = reshape(shape = concat_1199, x = matmul_119_cast_fp16)[name = string("reshape_359_cast_fp16")]; tensor transpose_2807_perm_0 = const()[name = string("transpose_2807_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2807 = transpose(perm = transpose_2807_perm_0, x = reshape_359_cast_fp16)[name = string("transpose_3864")]; tensor w_479_cast_fp16 = add(x = transpose_2807, y = transpose_2305)[name = string("w_479_cast_fp16")]; tensor var_2449_cast_fp16 = softmax(axis = var_2281, x = w_479_cast_fp16)[name = string("op_2449_cast_fp16")]; string var_2451_equation_0 = const()[name = string("op_2451_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2451_cast_fp16 = einsum(equation = var_2451_equation_0, values = (var_2371_cast_fp16_7, var_2449_cast_fp16))[name = string("op_2451_cast_fp16")]; tensor transpose_240_perm_0 = const()[name = string("transpose_240_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1204 = const()[name = string("concat_1204"), val = tensor([1, 104, 64])]; tensor transpose_240_cast_fp16 = transpose(perm = transpose_240_perm_0, x = var_2337_cast_fp16_8)[name = string("transpose_3863")]; tensor reshape_360_cast_fp16 = reshape(shape = concat_1204, x = transpose_240_cast_fp16)[name = string("reshape_360_cast_fp16")]; tensor transpose_241_perm_0 = const()[name = string("transpose_241_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1205 = const()[name = string("concat_1205"), val = tensor([1, 64, 104])]; tensor transpose_241_cast_fp16 = transpose(perm = transpose_241_perm_0, x = var_2354_cast_fp16_8)[name = string("transpose_3862")]; tensor reshape_361_cast_fp16 = reshape(shape = concat_1205, x = transpose_241_cast_fp16)[name = string("reshape_361_cast_fp16")]; bool matmul_120_transpose_x_0 = const()[name = string("matmul_120_transpose_x_0"), val = bool(false)]; bool matmul_120_transpose_y_0 = const()[name = string("matmul_120_transpose_y_0"), val = bool(false)]; tensor matmul_120_cast_fp16 = matmul(transpose_x = matmul_120_transpose_x_0, transpose_y = matmul_120_transpose_y_0, x = reshape_360_cast_fp16, y = reshape_361_cast_fp16)[name = string("matmul_120_cast_fp16")]; tensor concat_1209 = const()[name = string("concat_1209"), val = tensor([1, 1, 104, 104])]; tensor reshape_362_cast_fp16 = reshape(shape = concat_1209, x = matmul_120_cast_fp16)[name = string("reshape_362_cast_fp16")]; tensor transpose_2808_perm_0 = const()[name = string("transpose_2808_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2808 = transpose(perm = transpose_2808_perm_0, x = reshape_362_cast_fp16)[name = string("transpose_3861")]; tensor w_483_cast_fp16 = add(x = transpose_2808, y = transpose_2305)[name = string("w_483_cast_fp16")]; tensor var_2457_cast_fp16 = softmax(axis = var_2281, x = w_483_cast_fp16)[name = string("op_2457_cast_fp16")]; string var_2459_equation_0 = const()[name = string("op_2459_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2459_cast_fp16 = einsum(equation = var_2459_equation_0, values = (var_2371_cast_fp16_8, var_2457_cast_fp16))[name = string("op_2459_cast_fp16")]; tensor transpose_242_perm_0 = const()[name = string("transpose_242_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1214 = const()[name = string("concat_1214"), val = tensor([1, 104, 64])]; tensor transpose_242_cast_fp16 = transpose(perm = transpose_242_perm_0, x = var_2337_cast_fp16_9)[name = string("transpose_3860")]; tensor reshape_363_cast_fp16 = reshape(shape = concat_1214, x = transpose_242_cast_fp16)[name = string("reshape_363_cast_fp16")]; tensor transpose_243_perm_0 = const()[name = string("transpose_243_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1215 = const()[name = string("concat_1215"), val = tensor([1, 64, 104])]; tensor transpose_243_cast_fp16 = transpose(perm = transpose_243_perm_0, x = var_2354_cast_fp16_9)[name = string("transpose_3859")]; tensor reshape_364_cast_fp16 = reshape(shape = concat_1215, x = transpose_243_cast_fp16)[name = string("reshape_364_cast_fp16")]; bool matmul_121_transpose_x_0 = const()[name = string("matmul_121_transpose_x_0"), val = bool(false)]; bool matmul_121_transpose_y_0 = const()[name = string("matmul_121_transpose_y_0"), val = bool(false)]; tensor matmul_121_cast_fp16 = matmul(transpose_x = matmul_121_transpose_x_0, transpose_y = matmul_121_transpose_y_0, x = reshape_363_cast_fp16, y = reshape_364_cast_fp16)[name = string("matmul_121_cast_fp16")]; tensor concat_1219 = const()[name = string("concat_1219"), val = tensor([1, 1, 104, 104])]; tensor reshape_365_cast_fp16 = reshape(shape = concat_1219, x = matmul_121_cast_fp16)[name = string("reshape_365_cast_fp16")]; tensor transpose_2809_perm_0 = const()[name = string("transpose_2809_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2809 = transpose(perm = transpose_2809_perm_0, x = reshape_365_cast_fp16)[name = string("transpose_3858")]; tensor w_487_cast_fp16 = add(x = transpose_2809, y = transpose_2305)[name = string("w_487_cast_fp16")]; tensor var_2465_cast_fp16 = softmax(axis = var_2281, x = w_487_cast_fp16)[name = string("op_2465_cast_fp16")]; string var_2467_equation_0 = const()[name = string("op_2467_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2467_cast_fp16 = einsum(equation = var_2467_equation_0, values = (var_2371_cast_fp16_9, var_2465_cast_fp16))[name = string("op_2467_cast_fp16")]; tensor transpose_244_perm_0 = const()[name = string("transpose_244_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1224 = const()[name = string("concat_1224"), val = tensor([1, 104, 64])]; tensor transpose_244_cast_fp16 = transpose(perm = transpose_244_perm_0, x = var_2337_cast_fp16_10)[name = string("transpose_3857")]; tensor reshape_366_cast_fp16 = reshape(shape = concat_1224, x = transpose_244_cast_fp16)[name = string("reshape_366_cast_fp16")]; tensor transpose_245_perm_0 = const()[name = string("transpose_245_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1225 = const()[name = string("concat_1225"), val = tensor([1, 64, 104])]; tensor transpose_245_cast_fp16 = transpose(perm = transpose_245_perm_0, x = var_2354_cast_fp16_10)[name = string("transpose_3856")]; tensor reshape_367_cast_fp16 = reshape(shape = concat_1225, x = transpose_245_cast_fp16)[name = string("reshape_367_cast_fp16")]; bool matmul_122_transpose_x_0 = const()[name = string("matmul_122_transpose_x_0"), val = bool(false)]; bool matmul_122_transpose_y_0 = const()[name = string("matmul_122_transpose_y_0"), val = bool(false)]; tensor matmul_122_cast_fp16 = matmul(transpose_x = matmul_122_transpose_x_0, transpose_y = matmul_122_transpose_y_0, x = reshape_366_cast_fp16, y = reshape_367_cast_fp16)[name = string("matmul_122_cast_fp16")]; tensor concat_1229 = const()[name = string("concat_1229"), val = tensor([1, 1, 104, 104])]; tensor reshape_368_cast_fp16 = reshape(shape = concat_1229, x = matmul_122_cast_fp16)[name = string("reshape_368_cast_fp16")]; tensor transpose_2810_perm_0 = const()[name = string("transpose_2810_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2810 = transpose(perm = transpose_2810_perm_0, x = reshape_368_cast_fp16)[name = string("transpose_3855")]; tensor w_491_cast_fp16 = add(x = transpose_2810, y = transpose_2305)[name = string("w_491_cast_fp16")]; tensor var_2473_cast_fp16 = softmax(axis = var_2281, x = w_491_cast_fp16)[name = string("op_2473_cast_fp16")]; string var_2475_equation_0 = const()[name = string("op_2475_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2475_cast_fp16 = einsum(equation = var_2475_equation_0, values = (var_2371_cast_fp16_10, var_2473_cast_fp16))[name = string("op_2475_cast_fp16")]; tensor transpose_246_perm_0 = const()[name = string("transpose_246_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1234 = const()[name = string("concat_1234"), val = tensor([1, 104, 64])]; tensor transpose_246_cast_fp16 = transpose(perm = transpose_246_perm_0, x = var_2337_cast_fp16_11)[name = string("transpose_3854")]; tensor reshape_369_cast_fp16 = reshape(shape = concat_1234, x = transpose_246_cast_fp16)[name = string("reshape_369_cast_fp16")]; tensor transpose_247_perm_0 = const()[name = string("transpose_247_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1235 = const()[name = string("concat_1235"), val = tensor([1, 64, 104])]; tensor transpose_247_cast_fp16 = transpose(perm = transpose_247_perm_0, x = var_2354_cast_fp16_11)[name = string("transpose_3853")]; tensor reshape_370_cast_fp16 = reshape(shape = concat_1235, x = transpose_247_cast_fp16)[name = string("reshape_370_cast_fp16")]; bool matmul_123_transpose_x_0 = const()[name = string("matmul_123_transpose_x_0"), val = bool(false)]; bool matmul_123_transpose_y_0 = const()[name = string("matmul_123_transpose_y_0"), val = bool(false)]; tensor matmul_123_cast_fp16 = matmul(transpose_x = matmul_123_transpose_x_0, transpose_y = matmul_123_transpose_y_0, x = reshape_369_cast_fp16, y = reshape_370_cast_fp16)[name = string("matmul_123_cast_fp16")]; tensor concat_1239 = const()[name = string("concat_1239"), val = tensor([1, 1, 104, 104])]; tensor reshape_371_cast_fp16 = reshape(shape = concat_1239, x = matmul_123_cast_fp16)[name = string("reshape_371_cast_fp16")]; tensor transpose_2811_perm_0 = const()[name = string("transpose_2811_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2811 = transpose(perm = transpose_2811_perm_0, x = reshape_371_cast_fp16)[name = string("transpose_3852")]; tensor w_495_cast_fp16 = add(x = transpose_2811, y = transpose_2305)[name = string("w_495_cast_fp16")]; tensor var_2481_cast_fp16 = softmax(axis = var_2281, x = w_495_cast_fp16)[name = string("op_2481_cast_fp16")]; string var_2483_equation_0 = const()[name = string("op_2483_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2483_cast_fp16 = einsum(equation = var_2483_equation_0, values = (var_2371_cast_fp16_11, var_2481_cast_fp16))[name = string("op_2483_cast_fp16")]; tensor transpose_248_perm_0 = const()[name = string("transpose_248_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1244 = const()[name = string("concat_1244"), val = tensor([1, 104, 64])]; tensor transpose_248_cast_fp16 = transpose(perm = transpose_248_perm_0, x = var_2337_cast_fp16_12)[name = string("transpose_3851")]; tensor reshape_372_cast_fp16 = reshape(shape = concat_1244, x = transpose_248_cast_fp16)[name = string("reshape_372_cast_fp16")]; tensor transpose_249_perm_0 = const()[name = string("transpose_249_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1245 = const()[name = string("concat_1245"), val = tensor([1, 64, 104])]; tensor transpose_249_cast_fp16 = transpose(perm = transpose_249_perm_0, x = var_2354_cast_fp16_12)[name = string("transpose_3850")]; tensor reshape_373_cast_fp16 = reshape(shape = concat_1245, x = transpose_249_cast_fp16)[name = string("reshape_373_cast_fp16")]; bool matmul_124_transpose_x_0 = const()[name = string("matmul_124_transpose_x_0"), val = bool(false)]; bool matmul_124_transpose_y_0 = const()[name = string("matmul_124_transpose_y_0"), val = bool(false)]; tensor matmul_124_cast_fp16 = matmul(transpose_x = matmul_124_transpose_x_0, transpose_y = matmul_124_transpose_y_0, x = reshape_372_cast_fp16, y = reshape_373_cast_fp16)[name = string("matmul_124_cast_fp16")]; tensor concat_1249 = const()[name = string("concat_1249"), val = tensor([1, 1, 104, 104])]; tensor reshape_374_cast_fp16 = reshape(shape = concat_1249, x = matmul_124_cast_fp16)[name = string("reshape_374_cast_fp16")]; tensor transpose_2812_perm_0 = const()[name = string("transpose_2812_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2812 = transpose(perm = transpose_2812_perm_0, x = reshape_374_cast_fp16)[name = string("transpose_3849")]; tensor w_499_cast_fp16 = add(x = transpose_2812, y = transpose_2305)[name = string("w_499_cast_fp16")]; tensor var_2489_cast_fp16 = softmax(axis = var_2281, x = w_499_cast_fp16)[name = string("op_2489_cast_fp16")]; string var_2491_equation_0 = const()[name = string("op_2491_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2491_cast_fp16 = einsum(equation = var_2491_equation_0, values = (var_2371_cast_fp16_12, var_2489_cast_fp16))[name = string("op_2491_cast_fp16")]; tensor transpose_250_perm_0 = const()[name = string("transpose_250_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1254 = const()[name = string("concat_1254"), val = tensor([1, 104, 64])]; tensor transpose_250_cast_fp16 = transpose(perm = transpose_250_perm_0, x = var_2337_cast_fp16_13)[name = string("transpose_3848")]; tensor reshape_375_cast_fp16 = reshape(shape = concat_1254, x = transpose_250_cast_fp16)[name = string("reshape_375_cast_fp16")]; tensor transpose_251_perm_0 = const()[name = string("transpose_251_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1255 = const()[name = string("concat_1255"), val = tensor([1, 64, 104])]; tensor transpose_251_cast_fp16 = transpose(perm = transpose_251_perm_0, x = var_2354_cast_fp16_13)[name = string("transpose_3847")]; tensor reshape_376_cast_fp16 = reshape(shape = concat_1255, x = transpose_251_cast_fp16)[name = string("reshape_376_cast_fp16")]; bool matmul_125_transpose_x_0 = const()[name = string("matmul_125_transpose_x_0"), val = bool(false)]; bool matmul_125_transpose_y_0 = const()[name = string("matmul_125_transpose_y_0"), val = bool(false)]; tensor matmul_125_cast_fp16 = matmul(transpose_x = matmul_125_transpose_x_0, transpose_y = matmul_125_transpose_y_0, x = reshape_375_cast_fp16, y = reshape_376_cast_fp16)[name = string("matmul_125_cast_fp16")]; tensor concat_1259 = const()[name = string("concat_1259"), val = tensor([1, 1, 104, 104])]; tensor reshape_377_cast_fp16 = reshape(shape = concat_1259, x = matmul_125_cast_fp16)[name = string("reshape_377_cast_fp16")]; tensor transpose_2813_perm_0 = const()[name = string("transpose_2813_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2813 = transpose(perm = transpose_2813_perm_0, x = reshape_377_cast_fp16)[name = string("transpose_3846")]; tensor w_503_cast_fp16 = add(x = transpose_2813, y = transpose_2305)[name = string("w_503_cast_fp16")]; tensor var_2497_cast_fp16 = softmax(axis = var_2281, x = w_503_cast_fp16)[name = string("op_2497_cast_fp16")]; string var_2499_equation_0 = const()[name = string("op_2499_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2499_cast_fp16 = einsum(equation = var_2499_equation_0, values = (var_2371_cast_fp16_13, var_2497_cast_fp16))[name = string("op_2499_cast_fp16")]; tensor transpose_252_perm_0 = const()[name = string("transpose_252_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1264 = const()[name = string("concat_1264"), val = tensor([1, 104, 64])]; tensor transpose_252_cast_fp16 = transpose(perm = transpose_252_perm_0, x = var_2337_cast_fp16_14)[name = string("transpose_3845")]; tensor reshape_378_cast_fp16 = reshape(shape = concat_1264, x = transpose_252_cast_fp16)[name = string("reshape_378_cast_fp16")]; tensor transpose_253_perm_0 = const()[name = string("transpose_253_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1265 = const()[name = string("concat_1265"), val = tensor([1, 64, 104])]; tensor transpose_253_cast_fp16 = transpose(perm = transpose_253_perm_0, x = var_2354_cast_fp16_14)[name = string("transpose_3844")]; tensor reshape_379_cast_fp16 = reshape(shape = concat_1265, x = transpose_253_cast_fp16)[name = string("reshape_379_cast_fp16")]; bool matmul_126_transpose_x_0 = const()[name = string("matmul_126_transpose_x_0"), val = bool(false)]; bool matmul_126_transpose_y_0 = const()[name = string("matmul_126_transpose_y_0"), val = bool(false)]; tensor matmul_126_cast_fp16 = matmul(transpose_x = matmul_126_transpose_x_0, transpose_y = matmul_126_transpose_y_0, x = reshape_378_cast_fp16, y = reshape_379_cast_fp16)[name = string("matmul_126_cast_fp16")]; tensor concat_1269 = const()[name = string("concat_1269"), val = tensor([1, 1, 104, 104])]; tensor reshape_380_cast_fp16 = reshape(shape = concat_1269, x = matmul_126_cast_fp16)[name = string("reshape_380_cast_fp16")]; tensor transpose_2814_perm_0 = const()[name = string("transpose_2814_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2814 = transpose(perm = transpose_2814_perm_0, x = reshape_380_cast_fp16)[name = string("transpose_3843")]; tensor w_507_cast_fp16 = add(x = transpose_2814, y = transpose_2305)[name = string("w_507_cast_fp16")]; tensor var_2505_cast_fp16 = softmax(axis = var_2281, x = w_507_cast_fp16)[name = string("op_2505_cast_fp16")]; string var_2507_equation_0 = const()[name = string("op_2507_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2507_cast_fp16 = einsum(equation = var_2507_equation_0, values = (var_2371_cast_fp16_14, var_2505_cast_fp16))[name = string("op_2507_cast_fp16")]; tensor transpose_254_perm_0 = const()[name = string("transpose_254_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1274 = const()[name = string("concat_1274"), val = tensor([1, 104, 64])]; tensor transpose_254_cast_fp16 = transpose(perm = transpose_254_perm_0, x = var_2337_cast_fp16_15)[name = string("transpose_3842")]; tensor reshape_381_cast_fp16 = reshape(shape = concat_1274, x = transpose_254_cast_fp16)[name = string("reshape_381_cast_fp16")]; tensor transpose_255_perm_0 = const()[name = string("transpose_255_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1275 = const()[name = string("concat_1275"), val = tensor([1, 64, 104])]; tensor transpose_255_cast_fp16 = transpose(perm = transpose_255_perm_0, x = var_2354_cast_fp16_15)[name = string("transpose_3841")]; tensor reshape_382_cast_fp16 = reshape(shape = concat_1275, x = transpose_255_cast_fp16)[name = string("reshape_382_cast_fp16")]; bool matmul_127_transpose_x_0 = const()[name = string("matmul_127_transpose_x_0"), val = bool(false)]; bool matmul_127_transpose_y_0 = const()[name = string("matmul_127_transpose_y_0"), val = bool(false)]; tensor matmul_127_cast_fp16 = matmul(transpose_x = matmul_127_transpose_x_0, transpose_y = matmul_127_transpose_y_0, x = reshape_381_cast_fp16, y = reshape_382_cast_fp16)[name = string("matmul_127_cast_fp16")]; tensor concat_1279 = const()[name = string("concat_1279"), val = tensor([1, 1, 104, 104])]; tensor reshape_383_cast_fp16 = reshape(shape = concat_1279, x = matmul_127_cast_fp16)[name = string("reshape_383_cast_fp16")]; tensor transpose_2815_perm_0 = const()[name = string("transpose_2815_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2815 = transpose(perm = transpose_2815_perm_0, x = reshape_383_cast_fp16)[name = string("transpose_3840")]; tensor w_511_cast_fp16 = add(x = transpose_2815, y = transpose_2305)[name = string("w_511_cast_fp16")]; tensor var_2513_cast_fp16 = softmax(axis = var_2281, x = w_511_cast_fp16)[name = string("op_2513_cast_fp16")]; string var_2515_equation_0 = const()[name = string("op_2515_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2515_cast_fp16 = einsum(equation = var_2515_equation_0, values = (var_2371_cast_fp16_15, var_2513_cast_fp16))[name = string("op_2515_cast_fp16")]; bool input_67_interleave_0 = const()[name = string("input_67_interleave_0"), val = bool(false)]; tensor input_67_cast_fp16 = concat(axis = var_2281, interleave = input_67_interleave_0, values = (var_2395_cast_fp16, var_2403_cast_fp16, var_2411_cast_fp16, var_2419_cast_fp16, var_2427_cast_fp16, var_2435_cast_fp16, var_2443_cast_fp16, var_2451_cast_fp16, var_2459_cast_fp16, var_2467_cast_fp16, var_2475_cast_fp16, var_2483_cast_fp16, var_2491_cast_fp16, var_2499_cast_fp16, var_2507_cast_fp16, var_2515_cast_fp16))[name = string("input_67_cast_fp16")]; string var_2524_pad_type_0 = const()[name = string("op_2524_pad_type_0"), val = string("valid")]; tensor var_2524_strides_0 = const()[name = string("op_2524_strides_0"), val = tensor([1, 1])]; tensor var_2524_pad_0 = const()[name = string("op_2524_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2524_dilations_0 = const()[name = string("op_2524_dilations_0"), val = tensor([1, 1])]; int32 var_2524_groups_0 = const()[name = string("op_2524_groups_0"), val = int32(1)]; tensor layers_7_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_7_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(206725568)))]; tensor layers_7_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_7_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(208822784)))]; tensor var_2524_cast_fp16 = conv(bias = layers_7_self_attn_out_proj_bias_to_fp16, dilations = var_2524_dilations_0, groups = var_2524_groups_0, pad = var_2524_pad_0, pad_type = var_2524_pad_type_0, strides = var_2524_strides_0, weight = layers_7_self_attn_out_proj_weight_to_fp16, x = input_67_cast_fp16)[name = string("op_2524_cast_fp16")]; tensor x_87_cast_fp16 = add(x = x_83_cast_fp16, y = var_2524_cast_fp16)[name = string("x_87_cast_fp16")]; tensor mu_31_axes_0 = const()[name = string("mu_31_axes_0"), val = tensor([1])]; bool mu_31_keep_dims_0 = const()[name = string("mu_31_keep_dims_0"), val = bool(true)]; tensor mu_31_cast_fp16 = reduce_mean(axes = mu_31_axes_0, keep_dims = mu_31_keep_dims_0, x = x_87_cast_fp16)[name = string("mu_31_cast_fp16")]; tensor var_2530_cast_fp16 = sub(x = x_87_cast_fp16, y = mu_31_cast_fp16)[name = string("op_2530_cast_fp16")]; fp16 var_2284_promoted_1_to_fp16 = const()[name = string("op_2284_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_2531_cast_fp16 = pow(x = var_2530_cast_fp16, y = var_2284_promoted_1_to_fp16)[name = string("op_2531_cast_fp16")]; tensor var_31_axes_0 = const()[name = string("var_31_axes_0"), val = tensor([1])]; bool var_31_keep_dims_0 = const()[name = string("var_31_keep_dims_0"), val = bool(true)]; tensor var_31_cast_fp16 = reduce_mean(axes = var_31_axes_0, keep_dims = var_31_keep_dims_0, x = var_2531_cast_fp16)[name = string("var_31_cast_fp16")]; fp16 var_2535_to_fp16 = const()[name = string("op_2535_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2536_cast_fp16 = add(x = var_31_cast_fp16, y = var_2535_to_fp16)[name = string("op_2536_cast_fp16")]; fp32 var_2537_epsilon_0 = const()[name = string("op_2537_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2537_cast_fp16 = rsqrt(epsilon = var_2537_epsilon_0, x = var_2536_cast_fp16)[name = string("op_2537_cast_fp16")]; tensor x_89_cast_fp16 = mul(x = var_2530_cast_fp16, y = var_2537_cast_fp16)[name = string("x_89_cast_fp16")]; tensor input_69_gamma_0_to_fp16 = const()[name = string("input_69_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(208824896)))]; tensor input_69_beta_0_to_fp16 = const()[name = string("input_69_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(208827008)))]; fp16 input_69_epsilon_0_to_fp16 = const()[name = string("input_69_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_69_cast_fp16 = batch_norm(beta = input_69_beta_0_to_fp16, epsilon = input_69_epsilon_0_to_fp16, gamma = input_69_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_89_cast_fp16)[name = string("input_69_cast_fp16")]; string x_91_pad_type_0 = const()[name = string("x_91_pad_type_0"), val = string("valid")]; tensor x_91_strides_0 = const()[name = string("x_91_strides_0"), val = tensor([1, 1])]; tensor x_91_pad_0 = const()[name = string("x_91_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_91_dilations_0 = const()[name = string("x_91_dilations_0"), val = tensor([1, 1])]; int32 x_91_groups_0 = const()[name = string("x_91_groups_0"), val = int32(1)]; tensor layers_7_fc1_weight_to_fp16 = const()[name = string("layers_7_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(208829120)))]; tensor layers_7_fc1_bias_to_fp16 = const()[name = string("layers_7_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(217217792)))]; tensor x_91_cast_fp16 = conv(bias = layers_7_fc1_bias_to_fp16, dilations = x_91_dilations_0, groups = x_91_groups_0, pad = x_91_pad_0, pad_type = x_91_pad_type_0, strides = x_91_strides_0, weight = layers_7_fc1_weight_to_fp16, x = input_69_cast_fp16)[name = string("x_91_cast_fp16")]; fp16 var_2552_to_fp16 = const()[name = string("op_2552_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_2553_cast_fp16 = mul(x = x_91_cast_fp16, y = var_2552_to_fp16)[name = string("op_2553_cast_fp16")]; tensor var_2554_cast_fp16 = mul(x = var_2553_cast_fp16, y = x_91_cast_fp16)[name = string("op_2554_cast_fp16")]; tensor var_2555_cast_fp16 = mul(x = var_2554_cast_fp16, y = x_91_cast_fp16)[name = string("op_2555_cast_fp16")]; tensor var_2556_cast_fp16 = add(x = x_91_cast_fp16, y = var_2555_cast_fp16)[name = string("op_2556_cast_fp16")]; fp16 var_2557_to_fp16 = const()[name = string("op_2557_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_21_cast_fp16 = mul(x = var_2556_cast_fp16, y = var_2557_to_fp16)[name = string("u_21_cast_fp16")]; fp16 var_2559_to_fp16 = const()[name = string("op_2559_to_fp16"), val = fp16(0x1p-1)]; tensor var_2560_cast_fp16 = mul(x = x_91_cast_fp16, y = var_2559_to_fp16)[name = string("op_2560_cast_fp16")]; tensor var_2561_cast_fp16 = tanh(x = u_21_cast_fp16)[name = string("op_2561_cast_fp16")]; fp16 var_2562_to_fp16 = const()[name = string("op_2562_to_fp16"), val = fp16(0x1p+0)]; tensor var_2563_cast_fp16 = add(x = var_2561_cast_fp16, y = var_2562_to_fp16)[name = string("op_2563_cast_fp16")]; tensor input_71_cast_fp16 = mul(x = var_2560_cast_fp16, y = var_2563_cast_fp16)[name = string("input_71_cast_fp16")]; string h_15_pad_type_0 = const()[name = string("h_15_pad_type_0"), val = string("valid")]; tensor h_15_strides_0 = const()[name = string("h_15_strides_0"), val = tensor([1, 1])]; tensor h_15_pad_0 = const()[name = string("h_15_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_15_dilations_0 = const()[name = string("h_15_dilations_0"), val = tensor([1, 1])]; int32 h_15_groups_0 = const()[name = string("h_15_groups_0"), val = int32(1)]; tensor layers_7_fc2_weight_to_fp16 = const()[name = string("layers_7_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(217226048)))]; tensor layers_7_fc2_bias_to_fp16 = const()[name = string("layers_7_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(225614720)))]; tensor h_15_cast_fp16 = conv(bias = layers_7_fc2_bias_to_fp16, dilations = h_15_dilations_0, groups = h_15_groups_0, pad = h_15_pad_0, pad_type = h_15_pad_type_0, strides = h_15_strides_0, weight = layers_7_fc2_weight_to_fp16, x = input_71_cast_fp16)[name = string("h_15_cast_fp16")]; tensor x_93_cast_fp16 = add(x = x_87_cast_fp16, y = h_15_cast_fp16)[name = string("x_93_cast_fp16")]; int32 var_2579 = const()[name = string("op_2579"), val = int32(1)]; tensor mu_33_axes_0 = const()[name = string("mu_33_axes_0"), val = tensor([1])]; bool mu_33_keep_dims_0 = const()[name = string("mu_33_keep_dims_0"), val = bool(true)]; tensor mu_33_cast_fp16 = reduce_mean(axes = mu_33_axes_0, keep_dims = mu_33_keep_dims_0, x = x_93_cast_fp16)[name = string("mu_33_cast_fp16")]; tensor var_2593_cast_fp16 = sub(x = x_93_cast_fp16, y = mu_33_cast_fp16)[name = string("op_2593_cast_fp16")]; fp16 var_2582_promoted_to_fp16 = const()[name = string("op_2582_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_2594_cast_fp16 = pow(x = var_2593_cast_fp16, y = var_2582_promoted_to_fp16)[name = string("op_2594_cast_fp16")]; tensor var_33_axes_0 = const()[name = string("var_33_axes_0"), val = tensor([1])]; bool var_33_keep_dims_0 = const()[name = string("var_33_keep_dims_0"), val = bool(true)]; tensor var_33_cast_fp16 = reduce_mean(axes = var_33_axes_0, keep_dims = var_33_keep_dims_0, x = var_2594_cast_fp16)[name = string("var_33_cast_fp16")]; fp16 var_2598_to_fp16 = const()[name = string("op_2598_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2599_cast_fp16 = add(x = var_33_cast_fp16, y = var_2598_to_fp16)[name = string("op_2599_cast_fp16")]; fp32 var_2600_epsilon_0 = const()[name = string("op_2600_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2600_cast_fp16 = rsqrt(epsilon = var_2600_epsilon_0, x = var_2599_cast_fp16)[name = string("op_2600_cast_fp16")]; tensor x_95_cast_fp16 = mul(x = var_2593_cast_fp16, y = var_2600_cast_fp16)[name = string("x_95_cast_fp16")]; tensor input_73_gamma_0_to_fp16 = const()[name = string("input_73_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(225616832)))]; tensor input_73_beta_0_to_fp16 = const()[name = string("input_73_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(225618944)))]; fp16 input_73_epsilon_0_to_fp16 = const()[name = string("input_73_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_73_cast_fp16 = batch_norm(beta = input_73_beta_0_to_fp16, epsilon = input_73_epsilon_0_to_fp16, gamma = input_73_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_95_cast_fp16)[name = string("input_73_cast_fp16")]; string var_2618_pad_type_0 = const()[name = string("op_2618_pad_type_0"), val = string("valid")]; tensor var_2618_strides_0 = const()[name = string("op_2618_strides_0"), val = tensor([1, 1])]; tensor var_2618_pad_0 = const()[name = string("op_2618_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2618_dilations_0 = const()[name = string("op_2618_dilations_0"), val = tensor([1, 1])]; int32 var_2618_groups_0 = const()[name = string("op_2618_groups_0"), val = int32(1)]; tensor var_2620_weight_0_to_fp16 = const()[name = string("op_2620_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(225621056)))]; tensor var_2620_bias_0_to_fp16 = const()[name = string("op_2620_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(227718272)))]; tensor var_2620_cast_fp16 = conv(bias = var_2620_bias_0_to_fp16, dilations = var_2618_dilations_0, groups = var_2618_groups_0, pad = var_2618_pad_0, pad_type = var_2618_pad_type_0, strides = var_2618_strides_0, weight = var_2620_weight_0_to_fp16, x = input_73_cast_fp16)[name = string("op_2620_cast_fp16")]; string var_2627_pad_type_0 = const()[name = string("op_2627_pad_type_0"), val = string("valid")]; tensor var_2627_strides_0 = const()[name = string("op_2627_strides_0"), val = tensor([1, 1])]; tensor var_2627_pad_0 = const()[name = string("op_2627_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2627_dilations_0 = const()[name = string("op_2627_dilations_0"), val = tensor([1, 1])]; int32 var_2627_groups_0 = const()[name = string("op_2627_groups_0"), val = int32(1)]; tensor layers_8_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_8_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(227720384)))]; tensor layers_8_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_8_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(229817600)))]; tensor var_2627_cast_fp16 = conv(bias = layers_8_self_attn_k_proj_bias_to_fp16, dilations = var_2627_dilations_0, groups = var_2627_groups_0, pad = var_2627_pad_0, pad_type = var_2627_pad_type_0, strides = var_2627_strides_0, weight = layers_8_self_attn_k_proj_weight_to_fp16, x = input_73_cast_fp16)[name = string("op_2627_cast_fp16")]; string var_2634_pad_type_0 = const()[name = string("op_2634_pad_type_0"), val = string("valid")]; tensor var_2634_strides_0 = const()[name = string("op_2634_strides_0"), val = tensor([1, 1])]; tensor var_2634_pad_0 = const()[name = string("op_2634_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2634_dilations_0 = const()[name = string("op_2634_dilations_0"), val = tensor([1, 1])]; int32 var_2634_groups_0 = const()[name = string("op_2634_groups_0"), val = int32(1)]; tensor layers_8_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_8_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(229819712)))]; tensor layers_8_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_8_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(231916928)))]; tensor var_2634_cast_fp16 = conv(bias = layers_8_self_attn_v_proj_bias_to_fp16, dilations = var_2634_dilations_0, groups = var_2634_groups_0, pad = var_2634_pad_0, pad_type = var_2634_pad_type_0, strides = var_2634_strides_0, weight = layers_8_self_attn_v_proj_weight_to_fp16, x = input_73_cast_fp16)[name = string("op_2634_cast_fp16")]; tensor tile_24 = const()[name = string("tile_24"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(231919040)))]; int32 var_2635_axis_0 = const()[name = string("op_2635_axis_0"), val = int32(1)]; tensor var_2635_cast_fp16_0, tensor var_2635_cast_fp16_1, tensor var_2635_cast_fp16_2, tensor var_2635_cast_fp16_3, tensor var_2635_cast_fp16_4, tensor var_2635_cast_fp16_5, tensor var_2635_cast_fp16_6, tensor var_2635_cast_fp16_7, tensor var_2635_cast_fp16_8, tensor var_2635_cast_fp16_9, tensor var_2635_cast_fp16_10, tensor var_2635_cast_fp16_11, tensor var_2635_cast_fp16_12, tensor var_2635_cast_fp16_13, tensor var_2635_cast_fp16_14, tensor var_2635_cast_fp16_15 = split(axis = var_2635_axis_0, split_sizes = tile_24, x = var_2620_cast_fp16)[name = string("op_2635_cast_fp16")]; tensor tile_25 = const()[name = string("tile_25"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(231919168)))]; int32 var_2652_axis_0 = const()[name = string("op_2652_axis_0"), val = int32(1)]; tensor var_2652_cast_fp16_0, tensor var_2652_cast_fp16_1, tensor var_2652_cast_fp16_2, tensor var_2652_cast_fp16_3, tensor var_2652_cast_fp16_4, tensor var_2652_cast_fp16_5, tensor var_2652_cast_fp16_6, tensor var_2652_cast_fp16_7, tensor var_2652_cast_fp16_8, tensor var_2652_cast_fp16_9, tensor var_2652_cast_fp16_10, tensor var_2652_cast_fp16_11, tensor var_2652_cast_fp16_12, tensor var_2652_cast_fp16_13, tensor var_2652_cast_fp16_14, tensor var_2652_cast_fp16_15 = split(axis = var_2652_axis_0, split_sizes = tile_25, x = var_2627_cast_fp16)[name = string("op_2652_cast_fp16")]; tensor tile_26 = const()[name = string("tile_26"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(231919296)))]; int32 var_2669_axis_0 = const()[name = string("op_2669_axis_0"), val = int32(1)]; tensor var_2669_cast_fp16_0, tensor var_2669_cast_fp16_1, tensor var_2669_cast_fp16_2, tensor var_2669_cast_fp16_3, tensor var_2669_cast_fp16_4, tensor var_2669_cast_fp16_5, tensor var_2669_cast_fp16_6, tensor var_2669_cast_fp16_7, tensor var_2669_cast_fp16_8, tensor var_2669_cast_fp16_9, tensor var_2669_cast_fp16_10, tensor var_2669_cast_fp16_11, tensor var_2669_cast_fp16_12, tensor var_2669_cast_fp16_13, tensor var_2669_cast_fp16_14, tensor var_2669_cast_fp16_15 = split(axis = var_2669_axis_0, split_sizes = tile_26, x = var_2634_cast_fp16)[name = string("op_2669_cast_fp16")]; tensor transpose_256_perm_0 = const()[name = string("transpose_256_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1284 = const()[name = string("concat_1284"), val = tensor([1, 104, 64])]; tensor transpose_256_cast_fp16 = transpose(perm = transpose_256_perm_0, x = var_2635_cast_fp16_0)[name = string("transpose_3839")]; tensor reshape_384_cast_fp16 = reshape(shape = concat_1284, x = transpose_256_cast_fp16)[name = string("reshape_384_cast_fp16")]; tensor transpose_257_perm_0 = const()[name = string("transpose_257_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1285 = const()[name = string("concat_1285"), val = tensor([1, 64, 104])]; tensor transpose_257_cast_fp16 = transpose(perm = transpose_257_perm_0, x = var_2652_cast_fp16_0)[name = string("transpose_3838")]; tensor reshape_385_cast_fp16 = reshape(shape = concat_1285, x = transpose_257_cast_fp16)[name = string("reshape_385_cast_fp16")]; bool matmul_128_transpose_x_0 = const()[name = string("matmul_128_transpose_x_0"), val = bool(false)]; bool matmul_128_transpose_y_0 = const()[name = string("matmul_128_transpose_y_0"), val = bool(false)]; tensor matmul_128_cast_fp16 = matmul(transpose_x = matmul_128_transpose_x_0, transpose_y = matmul_128_transpose_y_0, x = reshape_384_cast_fp16, y = reshape_385_cast_fp16)[name = string("matmul_128_cast_fp16")]; tensor concat_1289 = const()[name = string("concat_1289"), val = tensor([1, 1, 104, 104])]; tensor reshape_386_cast_fp16 = reshape(shape = concat_1289, x = matmul_128_cast_fp16)[name = string("reshape_386_cast_fp16")]; tensor transpose_2816_perm_0 = const()[name = string("transpose_2816_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2816 = transpose(perm = transpose_2816_perm_0, x = reshape_386_cast_fp16)[name = string("transpose_3837")]; tensor w_515_cast_fp16 = add(x = transpose_2816, y = transpose_2305)[name = string("w_515_cast_fp16")]; tensor var_2691_cast_fp16 = softmax(axis = var_2579, x = w_515_cast_fp16)[name = string("op_2691_cast_fp16")]; string var_2693_equation_0 = const()[name = string("op_2693_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2693_cast_fp16 = einsum(equation = var_2693_equation_0, values = (var_2669_cast_fp16_0, var_2691_cast_fp16))[name = string("op_2693_cast_fp16")]; tensor transpose_258_perm_0 = const()[name = string("transpose_258_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1294 = const()[name = string("concat_1294"), val = tensor([1, 104, 64])]; tensor transpose_258_cast_fp16 = transpose(perm = transpose_258_perm_0, x = var_2635_cast_fp16_1)[name = string("transpose_3836")]; tensor reshape_387_cast_fp16 = reshape(shape = concat_1294, x = transpose_258_cast_fp16)[name = string("reshape_387_cast_fp16")]; tensor transpose_259_perm_0 = const()[name = string("transpose_259_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1295 = const()[name = string("concat_1295"), val = tensor([1, 64, 104])]; tensor transpose_259_cast_fp16 = transpose(perm = transpose_259_perm_0, x = var_2652_cast_fp16_1)[name = string("transpose_3835")]; tensor reshape_388_cast_fp16 = reshape(shape = concat_1295, x = transpose_259_cast_fp16)[name = string("reshape_388_cast_fp16")]; bool matmul_129_transpose_x_0 = const()[name = string("matmul_129_transpose_x_0"), val = bool(false)]; bool matmul_129_transpose_y_0 = const()[name = string("matmul_129_transpose_y_0"), val = bool(false)]; tensor matmul_129_cast_fp16 = matmul(transpose_x = matmul_129_transpose_x_0, transpose_y = matmul_129_transpose_y_0, x = reshape_387_cast_fp16, y = reshape_388_cast_fp16)[name = string("matmul_129_cast_fp16")]; tensor concat_1299 = const()[name = string("concat_1299"), val = tensor([1, 1, 104, 104])]; tensor reshape_389_cast_fp16 = reshape(shape = concat_1299, x = matmul_129_cast_fp16)[name = string("reshape_389_cast_fp16")]; tensor transpose_2817_perm_0 = const()[name = string("transpose_2817_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2817 = transpose(perm = transpose_2817_perm_0, x = reshape_389_cast_fp16)[name = string("transpose_3834")]; tensor w_519_cast_fp16 = add(x = transpose_2817, y = transpose_2305)[name = string("w_519_cast_fp16")]; tensor var_2699_cast_fp16 = softmax(axis = var_2579, x = w_519_cast_fp16)[name = string("op_2699_cast_fp16")]; string var_2701_equation_0 = const()[name = string("op_2701_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2701_cast_fp16 = einsum(equation = var_2701_equation_0, values = (var_2669_cast_fp16_1, var_2699_cast_fp16))[name = string("op_2701_cast_fp16")]; tensor transpose_260_perm_0 = const()[name = string("transpose_260_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1304 = const()[name = string("concat_1304"), val = tensor([1, 104, 64])]; tensor transpose_260_cast_fp16 = transpose(perm = transpose_260_perm_0, x = var_2635_cast_fp16_2)[name = string("transpose_3833")]; tensor reshape_390_cast_fp16 = reshape(shape = concat_1304, x = transpose_260_cast_fp16)[name = string("reshape_390_cast_fp16")]; tensor transpose_261_perm_0 = const()[name = string("transpose_261_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1305 = const()[name = string("concat_1305"), val = tensor([1, 64, 104])]; tensor transpose_261_cast_fp16 = transpose(perm = transpose_261_perm_0, x = var_2652_cast_fp16_2)[name = string("transpose_3832")]; tensor reshape_391_cast_fp16 = reshape(shape = concat_1305, x = transpose_261_cast_fp16)[name = string("reshape_391_cast_fp16")]; bool matmul_130_transpose_x_0 = const()[name = string("matmul_130_transpose_x_0"), val = bool(false)]; bool matmul_130_transpose_y_0 = const()[name = string("matmul_130_transpose_y_0"), val = bool(false)]; tensor matmul_130_cast_fp16 = matmul(transpose_x = matmul_130_transpose_x_0, transpose_y = matmul_130_transpose_y_0, x = reshape_390_cast_fp16, y = reshape_391_cast_fp16)[name = string("matmul_130_cast_fp16")]; tensor concat_1309 = const()[name = string("concat_1309"), val = tensor([1, 1, 104, 104])]; tensor reshape_392_cast_fp16 = reshape(shape = concat_1309, x = matmul_130_cast_fp16)[name = string("reshape_392_cast_fp16")]; tensor transpose_2818_perm_0 = const()[name = string("transpose_2818_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2818 = transpose(perm = transpose_2818_perm_0, x = reshape_392_cast_fp16)[name = string("transpose_3831")]; tensor w_523_cast_fp16 = add(x = transpose_2818, y = transpose_2305)[name = string("w_523_cast_fp16")]; tensor var_2707_cast_fp16 = softmax(axis = var_2579, x = w_523_cast_fp16)[name = string("op_2707_cast_fp16")]; string var_2709_equation_0 = const()[name = string("op_2709_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2709_cast_fp16 = einsum(equation = var_2709_equation_0, values = (var_2669_cast_fp16_2, var_2707_cast_fp16))[name = string("op_2709_cast_fp16")]; tensor transpose_262_perm_0 = const()[name = string("transpose_262_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1314 = const()[name = string("concat_1314"), val = tensor([1, 104, 64])]; tensor transpose_262_cast_fp16 = transpose(perm = transpose_262_perm_0, x = var_2635_cast_fp16_3)[name = string("transpose_3830")]; tensor reshape_393_cast_fp16 = reshape(shape = concat_1314, x = transpose_262_cast_fp16)[name = string("reshape_393_cast_fp16")]; tensor transpose_263_perm_0 = const()[name = string("transpose_263_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1315 = const()[name = string("concat_1315"), val = tensor([1, 64, 104])]; tensor transpose_263_cast_fp16 = transpose(perm = transpose_263_perm_0, x = var_2652_cast_fp16_3)[name = string("transpose_3829")]; tensor reshape_394_cast_fp16 = reshape(shape = concat_1315, x = transpose_263_cast_fp16)[name = string("reshape_394_cast_fp16")]; bool matmul_131_transpose_x_0 = const()[name = string("matmul_131_transpose_x_0"), val = bool(false)]; bool matmul_131_transpose_y_0 = const()[name = string("matmul_131_transpose_y_0"), val = bool(false)]; tensor matmul_131_cast_fp16 = matmul(transpose_x = matmul_131_transpose_x_0, transpose_y = matmul_131_transpose_y_0, x = reshape_393_cast_fp16, y = reshape_394_cast_fp16)[name = string("matmul_131_cast_fp16")]; tensor concat_1319 = const()[name = string("concat_1319"), val = tensor([1, 1, 104, 104])]; tensor reshape_395_cast_fp16 = reshape(shape = concat_1319, x = matmul_131_cast_fp16)[name = string("reshape_395_cast_fp16")]; tensor transpose_2819_perm_0 = const()[name = string("transpose_2819_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2819 = transpose(perm = transpose_2819_perm_0, x = reshape_395_cast_fp16)[name = string("transpose_3828")]; tensor w_527_cast_fp16 = add(x = transpose_2819, y = transpose_2305)[name = string("w_527_cast_fp16")]; tensor var_2715_cast_fp16 = softmax(axis = var_2579, x = w_527_cast_fp16)[name = string("op_2715_cast_fp16")]; string var_2717_equation_0 = const()[name = string("op_2717_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2717_cast_fp16 = einsum(equation = var_2717_equation_0, values = (var_2669_cast_fp16_3, var_2715_cast_fp16))[name = string("op_2717_cast_fp16")]; tensor transpose_264_perm_0 = const()[name = string("transpose_264_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1324 = const()[name = string("concat_1324"), val = tensor([1, 104, 64])]; tensor transpose_264_cast_fp16 = transpose(perm = transpose_264_perm_0, x = var_2635_cast_fp16_4)[name = string("transpose_3827")]; tensor reshape_396_cast_fp16 = reshape(shape = concat_1324, x = transpose_264_cast_fp16)[name = string("reshape_396_cast_fp16")]; tensor transpose_265_perm_0 = const()[name = string("transpose_265_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1325 = const()[name = string("concat_1325"), val = tensor([1, 64, 104])]; tensor transpose_265_cast_fp16 = transpose(perm = transpose_265_perm_0, x = var_2652_cast_fp16_4)[name = string("transpose_3826")]; tensor reshape_397_cast_fp16 = reshape(shape = concat_1325, x = transpose_265_cast_fp16)[name = string("reshape_397_cast_fp16")]; bool matmul_132_transpose_x_0 = const()[name = string("matmul_132_transpose_x_0"), val = bool(false)]; bool matmul_132_transpose_y_0 = const()[name = string("matmul_132_transpose_y_0"), val = bool(false)]; tensor matmul_132_cast_fp16 = matmul(transpose_x = matmul_132_transpose_x_0, transpose_y = matmul_132_transpose_y_0, x = reshape_396_cast_fp16, y = reshape_397_cast_fp16)[name = string("matmul_132_cast_fp16")]; tensor concat_1329 = const()[name = string("concat_1329"), val = tensor([1, 1, 104, 104])]; tensor reshape_398_cast_fp16 = reshape(shape = concat_1329, x = matmul_132_cast_fp16)[name = string("reshape_398_cast_fp16")]; tensor transpose_2820_perm_0 = const()[name = string("transpose_2820_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2820 = transpose(perm = transpose_2820_perm_0, x = reshape_398_cast_fp16)[name = string("transpose_3825")]; tensor w_531_cast_fp16 = add(x = transpose_2820, y = transpose_2305)[name = string("w_531_cast_fp16")]; tensor var_2723_cast_fp16 = softmax(axis = var_2579, x = w_531_cast_fp16)[name = string("op_2723_cast_fp16")]; string var_2725_equation_0 = const()[name = string("op_2725_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2725_cast_fp16 = einsum(equation = var_2725_equation_0, values = (var_2669_cast_fp16_4, var_2723_cast_fp16))[name = string("op_2725_cast_fp16")]; tensor transpose_266_perm_0 = const()[name = string("transpose_266_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1334 = const()[name = string("concat_1334"), val = tensor([1, 104, 64])]; tensor transpose_266_cast_fp16 = transpose(perm = transpose_266_perm_0, x = var_2635_cast_fp16_5)[name = string("transpose_3824")]; tensor reshape_399_cast_fp16 = reshape(shape = concat_1334, x = transpose_266_cast_fp16)[name = string("reshape_399_cast_fp16")]; tensor transpose_267_perm_0 = const()[name = string("transpose_267_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1335 = const()[name = string("concat_1335"), val = tensor([1, 64, 104])]; tensor transpose_267_cast_fp16 = transpose(perm = transpose_267_perm_0, x = var_2652_cast_fp16_5)[name = string("transpose_3823")]; tensor reshape_400_cast_fp16 = reshape(shape = concat_1335, x = transpose_267_cast_fp16)[name = string("reshape_400_cast_fp16")]; bool matmul_133_transpose_x_0 = const()[name = string("matmul_133_transpose_x_0"), val = bool(false)]; bool matmul_133_transpose_y_0 = const()[name = string("matmul_133_transpose_y_0"), val = bool(false)]; tensor matmul_133_cast_fp16 = matmul(transpose_x = matmul_133_transpose_x_0, transpose_y = matmul_133_transpose_y_0, x = reshape_399_cast_fp16, y = reshape_400_cast_fp16)[name = string("matmul_133_cast_fp16")]; tensor concat_1339 = const()[name = string("concat_1339"), val = tensor([1, 1, 104, 104])]; tensor reshape_401_cast_fp16 = reshape(shape = concat_1339, x = matmul_133_cast_fp16)[name = string("reshape_401_cast_fp16")]; tensor transpose_2821_perm_0 = const()[name = string("transpose_2821_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2821 = transpose(perm = transpose_2821_perm_0, x = reshape_401_cast_fp16)[name = string("transpose_3822")]; tensor w_535_cast_fp16 = add(x = transpose_2821, y = transpose_2305)[name = string("w_535_cast_fp16")]; tensor var_2731_cast_fp16 = softmax(axis = var_2579, x = w_535_cast_fp16)[name = string("op_2731_cast_fp16")]; string var_2733_equation_0 = const()[name = string("op_2733_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2733_cast_fp16 = einsum(equation = var_2733_equation_0, values = (var_2669_cast_fp16_5, var_2731_cast_fp16))[name = string("op_2733_cast_fp16")]; tensor transpose_268_perm_0 = const()[name = string("transpose_268_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1344 = const()[name = string("concat_1344"), val = tensor([1, 104, 64])]; tensor transpose_268_cast_fp16 = transpose(perm = transpose_268_perm_0, x = var_2635_cast_fp16_6)[name = string("transpose_3821")]; tensor reshape_402_cast_fp16 = reshape(shape = concat_1344, x = transpose_268_cast_fp16)[name = string("reshape_402_cast_fp16")]; tensor transpose_269_perm_0 = const()[name = string("transpose_269_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1345 = const()[name = string("concat_1345"), val = tensor([1, 64, 104])]; tensor transpose_269_cast_fp16 = transpose(perm = transpose_269_perm_0, x = var_2652_cast_fp16_6)[name = string("transpose_3820")]; tensor reshape_403_cast_fp16 = reshape(shape = concat_1345, x = transpose_269_cast_fp16)[name = string("reshape_403_cast_fp16")]; bool matmul_134_transpose_x_0 = const()[name = string("matmul_134_transpose_x_0"), val = bool(false)]; bool matmul_134_transpose_y_0 = const()[name = string("matmul_134_transpose_y_0"), val = bool(false)]; tensor matmul_134_cast_fp16 = matmul(transpose_x = matmul_134_transpose_x_0, transpose_y = matmul_134_transpose_y_0, x = reshape_402_cast_fp16, y = reshape_403_cast_fp16)[name = string("matmul_134_cast_fp16")]; tensor concat_1349 = const()[name = string("concat_1349"), val = tensor([1, 1, 104, 104])]; tensor reshape_404_cast_fp16 = reshape(shape = concat_1349, x = matmul_134_cast_fp16)[name = string("reshape_404_cast_fp16")]; tensor transpose_2822_perm_0 = const()[name = string("transpose_2822_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2822 = transpose(perm = transpose_2822_perm_0, x = reshape_404_cast_fp16)[name = string("transpose_3819")]; tensor w_539_cast_fp16 = add(x = transpose_2822, y = transpose_2305)[name = string("w_539_cast_fp16")]; tensor var_2739_cast_fp16 = softmax(axis = var_2579, x = w_539_cast_fp16)[name = string("op_2739_cast_fp16")]; string var_2741_equation_0 = const()[name = string("op_2741_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2741_cast_fp16 = einsum(equation = var_2741_equation_0, values = (var_2669_cast_fp16_6, var_2739_cast_fp16))[name = string("op_2741_cast_fp16")]; tensor transpose_270_perm_0 = const()[name = string("transpose_270_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1354 = const()[name = string("concat_1354"), val = tensor([1, 104, 64])]; tensor transpose_270_cast_fp16 = transpose(perm = transpose_270_perm_0, x = var_2635_cast_fp16_7)[name = string("transpose_3818")]; tensor reshape_405_cast_fp16 = reshape(shape = concat_1354, x = transpose_270_cast_fp16)[name = string("reshape_405_cast_fp16")]; tensor transpose_271_perm_0 = const()[name = string("transpose_271_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1355 = const()[name = string("concat_1355"), val = tensor([1, 64, 104])]; tensor transpose_271_cast_fp16 = transpose(perm = transpose_271_perm_0, x = var_2652_cast_fp16_7)[name = string("transpose_3817")]; tensor reshape_406_cast_fp16 = reshape(shape = concat_1355, x = transpose_271_cast_fp16)[name = string("reshape_406_cast_fp16")]; bool matmul_135_transpose_x_0 = const()[name = string("matmul_135_transpose_x_0"), val = bool(false)]; bool matmul_135_transpose_y_0 = const()[name = string("matmul_135_transpose_y_0"), val = bool(false)]; tensor matmul_135_cast_fp16 = matmul(transpose_x = matmul_135_transpose_x_0, transpose_y = matmul_135_transpose_y_0, x = reshape_405_cast_fp16, y = reshape_406_cast_fp16)[name = string("matmul_135_cast_fp16")]; tensor concat_1359 = const()[name = string("concat_1359"), val = tensor([1, 1, 104, 104])]; tensor reshape_407_cast_fp16 = reshape(shape = concat_1359, x = matmul_135_cast_fp16)[name = string("reshape_407_cast_fp16")]; tensor transpose_2823_perm_0 = const()[name = string("transpose_2823_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2823 = transpose(perm = transpose_2823_perm_0, x = reshape_407_cast_fp16)[name = string("transpose_3816")]; tensor w_543_cast_fp16 = add(x = transpose_2823, y = transpose_2305)[name = string("w_543_cast_fp16")]; tensor var_2747_cast_fp16 = softmax(axis = var_2579, x = w_543_cast_fp16)[name = string("op_2747_cast_fp16")]; string var_2749_equation_0 = const()[name = string("op_2749_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2749_cast_fp16 = einsum(equation = var_2749_equation_0, values = (var_2669_cast_fp16_7, var_2747_cast_fp16))[name = string("op_2749_cast_fp16")]; tensor transpose_272_perm_0 = const()[name = string("transpose_272_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1364 = const()[name = string("concat_1364"), val = tensor([1, 104, 64])]; tensor transpose_272_cast_fp16 = transpose(perm = transpose_272_perm_0, x = var_2635_cast_fp16_8)[name = string("transpose_3815")]; tensor reshape_408_cast_fp16 = reshape(shape = concat_1364, x = transpose_272_cast_fp16)[name = string("reshape_408_cast_fp16")]; tensor transpose_273_perm_0 = const()[name = string("transpose_273_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1365 = const()[name = string("concat_1365"), val = tensor([1, 64, 104])]; tensor transpose_273_cast_fp16 = transpose(perm = transpose_273_perm_0, x = var_2652_cast_fp16_8)[name = string("transpose_3814")]; tensor reshape_409_cast_fp16 = reshape(shape = concat_1365, x = transpose_273_cast_fp16)[name = string("reshape_409_cast_fp16")]; bool matmul_136_transpose_x_0 = const()[name = string("matmul_136_transpose_x_0"), val = bool(false)]; bool matmul_136_transpose_y_0 = const()[name = string("matmul_136_transpose_y_0"), val = bool(false)]; tensor matmul_136_cast_fp16 = matmul(transpose_x = matmul_136_transpose_x_0, transpose_y = matmul_136_transpose_y_0, x = reshape_408_cast_fp16, y = reshape_409_cast_fp16)[name = string("matmul_136_cast_fp16")]; tensor concat_1369 = const()[name = string("concat_1369"), val = tensor([1, 1, 104, 104])]; tensor reshape_410_cast_fp16 = reshape(shape = concat_1369, x = matmul_136_cast_fp16)[name = string("reshape_410_cast_fp16")]; tensor transpose_2824_perm_0 = const()[name = string("transpose_2824_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2824 = transpose(perm = transpose_2824_perm_0, x = reshape_410_cast_fp16)[name = string("transpose_3813")]; tensor w_547_cast_fp16 = add(x = transpose_2824, y = transpose_2305)[name = string("w_547_cast_fp16")]; tensor var_2755_cast_fp16 = softmax(axis = var_2579, x = w_547_cast_fp16)[name = string("op_2755_cast_fp16")]; string var_2757_equation_0 = const()[name = string("op_2757_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2757_cast_fp16 = einsum(equation = var_2757_equation_0, values = (var_2669_cast_fp16_8, var_2755_cast_fp16))[name = string("op_2757_cast_fp16")]; tensor transpose_274_perm_0 = const()[name = string("transpose_274_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1374 = const()[name = string("concat_1374"), val = tensor([1, 104, 64])]; tensor transpose_274_cast_fp16 = transpose(perm = transpose_274_perm_0, x = var_2635_cast_fp16_9)[name = string("transpose_3812")]; tensor reshape_411_cast_fp16 = reshape(shape = concat_1374, x = transpose_274_cast_fp16)[name = string("reshape_411_cast_fp16")]; tensor transpose_275_perm_0 = const()[name = string("transpose_275_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1375 = const()[name = string("concat_1375"), val = tensor([1, 64, 104])]; tensor transpose_275_cast_fp16 = transpose(perm = transpose_275_perm_0, x = var_2652_cast_fp16_9)[name = string("transpose_3811")]; tensor reshape_412_cast_fp16 = reshape(shape = concat_1375, x = transpose_275_cast_fp16)[name = string("reshape_412_cast_fp16")]; bool matmul_137_transpose_x_0 = const()[name = string("matmul_137_transpose_x_0"), val = bool(false)]; bool matmul_137_transpose_y_0 = const()[name = string("matmul_137_transpose_y_0"), val = bool(false)]; tensor matmul_137_cast_fp16 = matmul(transpose_x = matmul_137_transpose_x_0, transpose_y = matmul_137_transpose_y_0, x = reshape_411_cast_fp16, y = reshape_412_cast_fp16)[name = string("matmul_137_cast_fp16")]; tensor concat_1379 = const()[name = string("concat_1379"), val = tensor([1, 1, 104, 104])]; tensor reshape_413_cast_fp16 = reshape(shape = concat_1379, x = matmul_137_cast_fp16)[name = string("reshape_413_cast_fp16")]; tensor transpose_2825_perm_0 = const()[name = string("transpose_2825_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2825 = transpose(perm = transpose_2825_perm_0, x = reshape_413_cast_fp16)[name = string("transpose_3810")]; tensor w_551_cast_fp16 = add(x = transpose_2825, y = transpose_2305)[name = string("w_551_cast_fp16")]; tensor var_2763_cast_fp16 = softmax(axis = var_2579, x = w_551_cast_fp16)[name = string("op_2763_cast_fp16")]; string var_2765_equation_0 = const()[name = string("op_2765_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2765_cast_fp16 = einsum(equation = var_2765_equation_0, values = (var_2669_cast_fp16_9, var_2763_cast_fp16))[name = string("op_2765_cast_fp16")]; tensor transpose_276_perm_0 = const()[name = string("transpose_276_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1384 = const()[name = string("concat_1384"), val = tensor([1, 104, 64])]; tensor transpose_276_cast_fp16 = transpose(perm = transpose_276_perm_0, x = var_2635_cast_fp16_10)[name = string("transpose_3809")]; tensor reshape_414_cast_fp16 = reshape(shape = concat_1384, x = transpose_276_cast_fp16)[name = string("reshape_414_cast_fp16")]; tensor transpose_277_perm_0 = const()[name = string("transpose_277_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1385 = const()[name = string("concat_1385"), val = tensor([1, 64, 104])]; tensor transpose_277_cast_fp16 = transpose(perm = transpose_277_perm_0, x = var_2652_cast_fp16_10)[name = string("transpose_3808")]; tensor reshape_415_cast_fp16 = reshape(shape = concat_1385, x = transpose_277_cast_fp16)[name = string("reshape_415_cast_fp16")]; bool matmul_138_transpose_x_0 = const()[name = string("matmul_138_transpose_x_0"), val = bool(false)]; bool matmul_138_transpose_y_0 = const()[name = string("matmul_138_transpose_y_0"), val = bool(false)]; tensor matmul_138_cast_fp16 = matmul(transpose_x = matmul_138_transpose_x_0, transpose_y = matmul_138_transpose_y_0, x = reshape_414_cast_fp16, y = reshape_415_cast_fp16)[name = string("matmul_138_cast_fp16")]; tensor concat_1389 = const()[name = string("concat_1389"), val = tensor([1, 1, 104, 104])]; tensor reshape_416_cast_fp16 = reshape(shape = concat_1389, x = matmul_138_cast_fp16)[name = string("reshape_416_cast_fp16")]; tensor transpose_2826_perm_0 = const()[name = string("transpose_2826_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2826 = transpose(perm = transpose_2826_perm_0, x = reshape_416_cast_fp16)[name = string("transpose_3807")]; tensor w_555_cast_fp16 = add(x = transpose_2826, y = transpose_2305)[name = string("w_555_cast_fp16")]; tensor var_2771_cast_fp16 = softmax(axis = var_2579, x = w_555_cast_fp16)[name = string("op_2771_cast_fp16")]; string var_2773_equation_0 = const()[name = string("op_2773_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2773_cast_fp16 = einsum(equation = var_2773_equation_0, values = (var_2669_cast_fp16_10, var_2771_cast_fp16))[name = string("op_2773_cast_fp16")]; tensor transpose_278_perm_0 = const()[name = string("transpose_278_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1394 = const()[name = string("concat_1394"), val = tensor([1, 104, 64])]; tensor transpose_278_cast_fp16 = transpose(perm = transpose_278_perm_0, x = var_2635_cast_fp16_11)[name = string("transpose_3806")]; tensor reshape_417_cast_fp16 = reshape(shape = concat_1394, x = transpose_278_cast_fp16)[name = string("reshape_417_cast_fp16")]; tensor transpose_279_perm_0 = const()[name = string("transpose_279_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1395 = const()[name = string("concat_1395"), val = tensor([1, 64, 104])]; tensor transpose_279_cast_fp16 = transpose(perm = transpose_279_perm_0, x = var_2652_cast_fp16_11)[name = string("transpose_3805")]; tensor reshape_418_cast_fp16 = reshape(shape = concat_1395, x = transpose_279_cast_fp16)[name = string("reshape_418_cast_fp16")]; bool matmul_139_transpose_x_0 = const()[name = string("matmul_139_transpose_x_0"), val = bool(false)]; bool matmul_139_transpose_y_0 = const()[name = string("matmul_139_transpose_y_0"), val = bool(false)]; tensor matmul_139_cast_fp16 = matmul(transpose_x = matmul_139_transpose_x_0, transpose_y = matmul_139_transpose_y_0, x = reshape_417_cast_fp16, y = reshape_418_cast_fp16)[name = string("matmul_139_cast_fp16")]; tensor concat_1399 = const()[name = string("concat_1399"), val = tensor([1, 1, 104, 104])]; tensor reshape_419_cast_fp16 = reshape(shape = concat_1399, x = matmul_139_cast_fp16)[name = string("reshape_419_cast_fp16")]; tensor transpose_2827_perm_0 = const()[name = string("transpose_2827_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2827 = transpose(perm = transpose_2827_perm_0, x = reshape_419_cast_fp16)[name = string("transpose_3804")]; tensor w_559_cast_fp16 = add(x = transpose_2827, y = transpose_2305)[name = string("w_559_cast_fp16")]; tensor var_2779_cast_fp16 = softmax(axis = var_2579, x = w_559_cast_fp16)[name = string("op_2779_cast_fp16")]; string var_2781_equation_0 = const()[name = string("op_2781_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2781_cast_fp16 = einsum(equation = var_2781_equation_0, values = (var_2669_cast_fp16_11, var_2779_cast_fp16))[name = string("op_2781_cast_fp16")]; tensor transpose_280_perm_0 = const()[name = string("transpose_280_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1404 = const()[name = string("concat_1404"), val = tensor([1, 104, 64])]; tensor transpose_280_cast_fp16 = transpose(perm = transpose_280_perm_0, x = var_2635_cast_fp16_12)[name = string("transpose_3803")]; tensor reshape_420_cast_fp16 = reshape(shape = concat_1404, x = transpose_280_cast_fp16)[name = string("reshape_420_cast_fp16")]; tensor transpose_281_perm_0 = const()[name = string("transpose_281_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1405 = const()[name = string("concat_1405"), val = tensor([1, 64, 104])]; tensor transpose_281_cast_fp16 = transpose(perm = transpose_281_perm_0, x = var_2652_cast_fp16_12)[name = string("transpose_3802")]; tensor reshape_421_cast_fp16 = reshape(shape = concat_1405, x = transpose_281_cast_fp16)[name = string("reshape_421_cast_fp16")]; bool matmul_140_transpose_x_0 = const()[name = string("matmul_140_transpose_x_0"), val = bool(false)]; bool matmul_140_transpose_y_0 = const()[name = string("matmul_140_transpose_y_0"), val = bool(false)]; tensor matmul_140_cast_fp16 = matmul(transpose_x = matmul_140_transpose_x_0, transpose_y = matmul_140_transpose_y_0, x = reshape_420_cast_fp16, y = reshape_421_cast_fp16)[name = string("matmul_140_cast_fp16")]; tensor concat_1409 = const()[name = string("concat_1409"), val = tensor([1, 1, 104, 104])]; tensor reshape_422_cast_fp16 = reshape(shape = concat_1409, x = matmul_140_cast_fp16)[name = string("reshape_422_cast_fp16")]; tensor transpose_2828_perm_0 = const()[name = string("transpose_2828_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2828 = transpose(perm = transpose_2828_perm_0, x = reshape_422_cast_fp16)[name = string("transpose_3801")]; tensor w_563_cast_fp16 = add(x = transpose_2828, y = transpose_2305)[name = string("w_563_cast_fp16")]; tensor var_2787_cast_fp16 = softmax(axis = var_2579, x = w_563_cast_fp16)[name = string("op_2787_cast_fp16")]; string var_2789_equation_0 = const()[name = string("op_2789_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2789_cast_fp16 = einsum(equation = var_2789_equation_0, values = (var_2669_cast_fp16_12, var_2787_cast_fp16))[name = string("op_2789_cast_fp16")]; tensor transpose_282_perm_0 = const()[name = string("transpose_282_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1414 = const()[name = string("concat_1414"), val = tensor([1, 104, 64])]; tensor transpose_282_cast_fp16 = transpose(perm = transpose_282_perm_0, x = var_2635_cast_fp16_13)[name = string("transpose_3800")]; tensor reshape_423_cast_fp16 = reshape(shape = concat_1414, x = transpose_282_cast_fp16)[name = string("reshape_423_cast_fp16")]; tensor transpose_283_perm_0 = const()[name = string("transpose_283_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1415 = const()[name = string("concat_1415"), val = tensor([1, 64, 104])]; tensor transpose_283_cast_fp16 = transpose(perm = transpose_283_perm_0, x = var_2652_cast_fp16_13)[name = string("transpose_3799")]; tensor reshape_424_cast_fp16 = reshape(shape = concat_1415, x = transpose_283_cast_fp16)[name = string("reshape_424_cast_fp16")]; bool matmul_141_transpose_x_0 = const()[name = string("matmul_141_transpose_x_0"), val = bool(false)]; bool matmul_141_transpose_y_0 = const()[name = string("matmul_141_transpose_y_0"), val = bool(false)]; tensor matmul_141_cast_fp16 = matmul(transpose_x = matmul_141_transpose_x_0, transpose_y = matmul_141_transpose_y_0, x = reshape_423_cast_fp16, y = reshape_424_cast_fp16)[name = string("matmul_141_cast_fp16")]; tensor concat_1419 = const()[name = string("concat_1419"), val = tensor([1, 1, 104, 104])]; tensor reshape_425_cast_fp16 = reshape(shape = concat_1419, x = matmul_141_cast_fp16)[name = string("reshape_425_cast_fp16")]; tensor transpose_2829_perm_0 = const()[name = string("transpose_2829_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2829 = transpose(perm = transpose_2829_perm_0, x = reshape_425_cast_fp16)[name = string("transpose_3798")]; tensor w_567_cast_fp16 = add(x = transpose_2829, y = transpose_2305)[name = string("w_567_cast_fp16")]; tensor var_2795_cast_fp16 = softmax(axis = var_2579, x = w_567_cast_fp16)[name = string("op_2795_cast_fp16")]; string var_2797_equation_0 = const()[name = string("op_2797_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2797_cast_fp16 = einsum(equation = var_2797_equation_0, values = (var_2669_cast_fp16_13, var_2795_cast_fp16))[name = string("op_2797_cast_fp16")]; tensor transpose_284_perm_0 = const()[name = string("transpose_284_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1424 = const()[name = string("concat_1424"), val = tensor([1, 104, 64])]; tensor transpose_284_cast_fp16 = transpose(perm = transpose_284_perm_0, x = var_2635_cast_fp16_14)[name = string("transpose_3797")]; tensor reshape_426_cast_fp16 = reshape(shape = concat_1424, x = transpose_284_cast_fp16)[name = string("reshape_426_cast_fp16")]; tensor transpose_285_perm_0 = const()[name = string("transpose_285_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1425 = const()[name = string("concat_1425"), val = tensor([1, 64, 104])]; tensor transpose_285_cast_fp16 = transpose(perm = transpose_285_perm_0, x = var_2652_cast_fp16_14)[name = string("transpose_3796")]; tensor reshape_427_cast_fp16 = reshape(shape = concat_1425, x = transpose_285_cast_fp16)[name = string("reshape_427_cast_fp16")]; bool matmul_142_transpose_x_0 = const()[name = string("matmul_142_transpose_x_0"), val = bool(false)]; bool matmul_142_transpose_y_0 = const()[name = string("matmul_142_transpose_y_0"), val = bool(false)]; tensor matmul_142_cast_fp16 = matmul(transpose_x = matmul_142_transpose_x_0, transpose_y = matmul_142_transpose_y_0, x = reshape_426_cast_fp16, y = reshape_427_cast_fp16)[name = string("matmul_142_cast_fp16")]; tensor concat_1429 = const()[name = string("concat_1429"), val = tensor([1, 1, 104, 104])]; tensor reshape_428_cast_fp16 = reshape(shape = concat_1429, x = matmul_142_cast_fp16)[name = string("reshape_428_cast_fp16")]; tensor transpose_2830_perm_0 = const()[name = string("transpose_2830_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2830 = transpose(perm = transpose_2830_perm_0, x = reshape_428_cast_fp16)[name = string("transpose_3795")]; tensor w_571_cast_fp16 = add(x = transpose_2830, y = transpose_2305)[name = string("w_571_cast_fp16")]; tensor var_2803_cast_fp16 = softmax(axis = var_2579, x = w_571_cast_fp16)[name = string("op_2803_cast_fp16")]; string var_2805_equation_0 = const()[name = string("op_2805_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2805_cast_fp16 = einsum(equation = var_2805_equation_0, values = (var_2669_cast_fp16_14, var_2803_cast_fp16))[name = string("op_2805_cast_fp16")]; tensor transpose_286_perm_0 = const()[name = string("transpose_286_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1434 = const()[name = string("concat_1434"), val = tensor([1, 104, 64])]; tensor transpose_286_cast_fp16 = transpose(perm = transpose_286_perm_0, x = var_2635_cast_fp16_15)[name = string("transpose_3794")]; tensor reshape_429_cast_fp16 = reshape(shape = concat_1434, x = transpose_286_cast_fp16)[name = string("reshape_429_cast_fp16")]; tensor transpose_287_perm_0 = const()[name = string("transpose_287_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1435 = const()[name = string("concat_1435"), val = tensor([1, 64, 104])]; tensor transpose_287_cast_fp16 = transpose(perm = transpose_287_perm_0, x = var_2652_cast_fp16_15)[name = string("transpose_3793")]; tensor reshape_430_cast_fp16 = reshape(shape = concat_1435, x = transpose_287_cast_fp16)[name = string("reshape_430_cast_fp16")]; bool matmul_143_transpose_x_0 = const()[name = string("matmul_143_transpose_x_0"), val = bool(false)]; bool matmul_143_transpose_y_0 = const()[name = string("matmul_143_transpose_y_0"), val = bool(false)]; tensor matmul_143_cast_fp16 = matmul(transpose_x = matmul_143_transpose_x_0, transpose_y = matmul_143_transpose_y_0, x = reshape_429_cast_fp16, y = reshape_430_cast_fp16)[name = string("matmul_143_cast_fp16")]; tensor concat_1439 = const()[name = string("concat_1439"), val = tensor([1, 1, 104, 104])]; tensor reshape_431_cast_fp16 = reshape(shape = concat_1439, x = matmul_143_cast_fp16)[name = string("reshape_431_cast_fp16")]; tensor transpose_2831_perm_0 = const()[name = string("transpose_2831_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2831 = transpose(perm = transpose_2831_perm_0, x = reshape_431_cast_fp16)[name = string("transpose_3792")]; tensor w_575_cast_fp16 = add(x = transpose_2831, y = transpose_2305)[name = string("w_575_cast_fp16")]; tensor var_2811_cast_fp16 = softmax(axis = var_2579, x = w_575_cast_fp16)[name = string("op_2811_cast_fp16")]; string var_2813_equation_0 = const()[name = string("op_2813_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2813_cast_fp16 = einsum(equation = var_2813_equation_0, values = (var_2669_cast_fp16_15, var_2811_cast_fp16))[name = string("op_2813_cast_fp16")]; bool input_75_interleave_0 = const()[name = string("input_75_interleave_0"), val = bool(false)]; tensor input_75_cast_fp16 = concat(axis = var_2579, interleave = input_75_interleave_0, values = (var_2693_cast_fp16, var_2701_cast_fp16, var_2709_cast_fp16, var_2717_cast_fp16, var_2725_cast_fp16, var_2733_cast_fp16, var_2741_cast_fp16, var_2749_cast_fp16, var_2757_cast_fp16, var_2765_cast_fp16, var_2773_cast_fp16, var_2781_cast_fp16, var_2789_cast_fp16, var_2797_cast_fp16, var_2805_cast_fp16, var_2813_cast_fp16))[name = string("input_75_cast_fp16")]; string var_2822_pad_type_0 = const()[name = string("op_2822_pad_type_0"), val = string("valid")]; tensor var_2822_strides_0 = const()[name = string("op_2822_strides_0"), val = tensor([1, 1])]; tensor var_2822_pad_0 = const()[name = string("op_2822_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2822_dilations_0 = const()[name = string("op_2822_dilations_0"), val = tensor([1, 1])]; int32 var_2822_groups_0 = const()[name = string("op_2822_groups_0"), val = int32(1)]; tensor layers_8_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_8_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(231919424)))]; tensor layers_8_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_8_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(234016640)))]; tensor var_2822_cast_fp16 = conv(bias = layers_8_self_attn_out_proj_bias_to_fp16, dilations = var_2822_dilations_0, groups = var_2822_groups_0, pad = var_2822_pad_0, pad_type = var_2822_pad_type_0, strides = var_2822_strides_0, weight = layers_8_self_attn_out_proj_weight_to_fp16, x = input_75_cast_fp16)[name = string("op_2822_cast_fp16")]; tensor x_97_cast_fp16 = add(x = x_93_cast_fp16, y = var_2822_cast_fp16)[name = string("x_97_cast_fp16")]; tensor mu_35_axes_0 = const()[name = string("mu_35_axes_0"), val = tensor([1])]; bool mu_35_keep_dims_0 = const()[name = string("mu_35_keep_dims_0"), val = bool(true)]; tensor mu_35_cast_fp16 = reduce_mean(axes = mu_35_axes_0, keep_dims = mu_35_keep_dims_0, x = x_97_cast_fp16)[name = string("mu_35_cast_fp16")]; tensor var_2828_cast_fp16 = sub(x = x_97_cast_fp16, y = mu_35_cast_fp16)[name = string("op_2828_cast_fp16")]; fp16 var_2582_promoted_1_to_fp16 = const()[name = string("op_2582_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_2829_cast_fp16 = pow(x = var_2828_cast_fp16, y = var_2582_promoted_1_to_fp16)[name = string("op_2829_cast_fp16")]; tensor var_35_axes_0 = const()[name = string("var_35_axes_0"), val = tensor([1])]; bool var_35_keep_dims_0 = const()[name = string("var_35_keep_dims_0"), val = bool(true)]; tensor var_35_cast_fp16 = reduce_mean(axes = var_35_axes_0, keep_dims = var_35_keep_dims_0, x = var_2829_cast_fp16)[name = string("var_35_cast_fp16")]; fp16 var_2833_to_fp16 = const()[name = string("op_2833_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2834_cast_fp16 = add(x = var_35_cast_fp16, y = var_2833_to_fp16)[name = string("op_2834_cast_fp16")]; fp32 var_2835_epsilon_0 = const()[name = string("op_2835_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2835_cast_fp16 = rsqrt(epsilon = var_2835_epsilon_0, x = var_2834_cast_fp16)[name = string("op_2835_cast_fp16")]; tensor x_99_cast_fp16 = mul(x = var_2828_cast_fp16, y = var_2835_cast_fp16)[name = string("x_99_cast_fp16")]; tensor input_77_gamma_0_to_fp16 = const()[name = string("input_77_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(234018752)))]; tensor input_77_beta_0_to_fp16 = const()[name = string("input_77_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(234020864)))]; fp16 input_77_epsilon_0_to_fp16 = const()[name = string("input_77_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_77_cast_fp16 = batch_norm(beta = input_77_beta_0_to_fp16, epsilon = input_77_epsilon_0_to_fp16, gamma = input_77_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_99_cast_fp16)[name = string("input_77_cast_fp16")]; string x_101_pad_type_0 = const()[name = string("x_101_pad_type_0"), val = string("valid")]; tensor x_101_strides_0 = const()[name = string("x_101_strides_0"), val = tensor([1, 1])]; tensor x_101_pad_0 = const()[name = string("x_101_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_101_dilations_0 = const()[name = string("x_101_dilations_0"), val = tensor([1, 1])]; int32 x_101_groups_0 = const()[name = string("x_101_groups_0"), val = int32(1)]; tensor layers_8_fc1_weight_to_fp16 = const()[name = string("layers_8_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(234022976)))]; tensor layers_8_fc1_bias_to_fp16 = const()[name = string("layers_8_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(242411648)))]; tensor x_101_cast_fp16 = conv(bias = layers_8_fc1_bias_to_fp16, dilations = x_101_dilations_0, groups = x_101_groups_0, pad = x_101_pad_0, pad_type = x_101_pad_type_0, strides = x_101_strides_0, weight = layers_8_fc1_weight_to_fp16, x = input_77_cast_fp16)[name = string("x_101_cast_fp16")]; fp16 var_2850_to_fp16 = const()[name = string("op_2850_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_2851_cast_fp16 = mul(x = x_101_cast_fp16, y = var_2850_to_fp16)[name = string("op_2851_cast_fp16")]; tensor var_2852_cast_fp16 = mul(x = var_2851_cast_fp16, y = x_101_cast_fp16)[name = string("op_2852_cast_fp16")]; tensor var_2853_cast_fp16 = mul(x = var_2852_cast_fp16, y = x_101_cast_fp16)[name = string("op_2853_cast_fp16")]; tensor var_2854_cast_fp16 = add(x = x_101_cast_fp16, y = var_2853_cast_fp16)[name = string("op_2854_cast_fp16")]; fp16 var_2855_to_fp16 = const()[name = string("op_2855_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_23_cast_fp16 = mul(x = var_2854_cast_fp16, y = var_2855_to_fp16)[name = string("u_23_cast_fp16")]; fp16 var_2857_to_fp16 = const()[name = string("op_2857_to_fp16"), val = fp16(0x1p-1)]; tensor var_2858_cast_fp16 = mul(x = x_101_cast_fp16, y = var_2857_to_fp16)[name = string("op_2858_cast_fp16")]; tensor var_2859_cast_fp16 = tanh(x = u_23_cast_fp16)[name = string("op_2859_cast_fp16")]; fp16 var_2860_to_fp16 = const()[name = string("op_2860_to_fp16"), val = fp16(0x1p+0)]; tensor var_2861_cast_fp16 = add(x = var_2859_cast_fp16, y = var_2860_to_fp16)[name = string("op_2861_cast_fp16")]; tensor input_79_cast_fp16 = mul(x = var_2858_cast_fp16, y = var_2861_cast_fp16)[name = string("input_79_cast_fp16")]; string h_17_pad_type_0 = const()[name = string("h_17_pad_type_0"), val = string("valid")]; tensor h_17_strides_0 = const()[name = string("h_17_strides_0"), val = tensor([1, 1])]; tensor h_17_pad_0 = const()[name = string("h_17_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_17_dilations_0 = const()[name = string("h_17_dilations_0"), val = tensor([1, 1])]; int32 h_17_groups_0 = const()[name = string("h_17_groups_0"), val = int32(1)]; tensor layers_8_fc2_weight_to_fp16 = const()[name = string("layers_8_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(242419904)))]; tensor layers_8_fc2_bias_to_fp16 = const()[name = string("layers_8_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(250808576)))]; tensor h_17_cast_fp16 = conv(bias = layers_8_fc2_bias_to_fp16, dilations = h_17_dilations_0, groups = h_17_groups_0, pad = h_17_pad_0, pad_type = h_17_pad_type_0, strides = h_17_strides_0, weight = layers_8_fc2_weight_to_fp16, x = input_79_cast_fp16)[name = string("h_17_cast_fp16")]; tensor x_103_cast_fp16 = add(x = x_97_cast_fp16, y = h_17_cast_fp16)[name = string("x_103_cast_fp16")]; int32 var_2877 = const()[name = string("op_2877"), val = int32(1)]; tensor mu_37_axes_0 = const()[name = string("mu_37_axes_0"), val = tensor([1])]; bool mu_37_keep_dims_0 = const()[name = string("mu_37_keep_dims_0"), val = bool(true)]; tensor mu_37_cast_fp16 = reduce_mean(axes = mu_37_axes_0, keep_dims = mu_37_keep_dims_0, x = x_103_cast_fp16)[name = string("mu_37_cast_fp16")]; tensor var_2891_cast_fp16 = sub(x = x_103_cast_fp16, y = mu_37_cast_fp16)[name = string("op_2891_cast_fp16")]; fp16 var_2880_promoted_to_fp16 = const()[name = string("op_2880_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_2892_cast_fp16 = pow(x = var_2891_cast_fp16, y = var_2880_promoted_to_fp16)[name = string("op_2892_cast_fp16")]; tensor var_37_axes_0 = const()[name = string("var_37_axes_0"), val = tensor([1])]; bool var_37_keep_dims_0 = const()[name = string("var_37_keep_dims_0"), val = bool(true)]; tensor var_37_cast_fp16 = reduce_mean(axes = var_37_axes_0, keep_dims = var_37_keep_dims_0, x = var_2892_cast_fp16)[name = string("var_37_cast_fp16")]; fp16 var_2896_to_fp16 = const()[name = string("op_2896_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_2897_cast_fp16 = add(x = var_37_cast_fp16, y = var_2896_to_fp16)[name = string("op_2897_cast_fp16")]; fp32 var_2898_epsilon_0 = const()[name = string("op_2898_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_2898_cast_fp16 = rsqrt(epsilon = var_2898_epsilon_0, x = var_2897_cast_fp16)[name = string("op_2898_cast_fp16")]; tensor x_105_cast_fp16 = mul(x = var_2891_cast_fp16, y = var_2898_cast_fp16)[name = string("x_105_cast_fp16")]; tensor input_81_gamma_0_to_fp16 = const()[name = string("input_81_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(250810688)))]; tensor input_81_beta_0_to_fp16 = const()[name = string("input_81_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(250812800)))]; fp16 input_81_epsilon_0_to_fp16 = const()[name = string("input_81_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_81_cast_fp16 = batch_norm(beta = input_81_beta_0_to_fp16, epsilon = input_81_epsilon_0_to_fp16, gamma = input_81_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_105_cast_fp16)[name = string("input_81_cast_fp16")]; string var_2916_pad_type_0 = const()[name = string("op_2916_pad_type_0"), val = string("valid")]; tensor var_2916_strides_0 = const()[name = string("op_2916_strides_0"), val = tensor([1, 1])]; tensor var_2916_pad_0 = const()[name = string("op_2916_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2916_dilations_0 = const()[name = string("op_2916_dilations_0"), val = tensor([1, 1])]; int32 var_2916_groups_0 = const()[name = string("op_2916_groups_0"), val = int32(1)]; tensor var_2918_weight_0_to_fp16 = const()[name = string("op_2918_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(250814912)))]; tensor var_2918_bias_0_to_fp16 = const()[name = string("op_2918_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(252912128)))]; tensor var_2918_cast_fp16 = conv(bias = var_2918_bias_0_to_fp16, dilations = var_2916_dilations_0, groups = var_2916_groups_0, pad = var_2916_pad_0, pad_type = var_2916_pad_type_0, strides = var_2916_strides_0, weight = var_2918_weight_0_to_fp16, x = input_81_cast_fp16)[name = string("op_2918_cast_fp16")]; string var_2925_pad_type_0 = const()[name = string("op_2925_pad_type_0"), val = string("valid")]; tensor var_2925_strides_0 = const()[name = string("op_2925_strides_0"), val = tensor([1, 1])]; tensor var_2925_pad_0 = const()[name = string("op_2925_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2925_dilations_0 = const()[name = string("op_2925_dilations_0"), val = tensor([1, 1])]; int32 var_2925_groups_0 = const()[name = string("op_2925_groups_0"), val = int32(1)]; tensor layers_9_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_9_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(252914240)))]; tensor layers_9_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_9_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(255011456)))]; tensor var_2925_cast_fp16 = conv(bias = layers_9_self_attn_k_proj_bias_to_fp16, dilations = var_2925_dilations_0, groups = var_2925_groups_0, pad = var_2925_pad_0, pad_type = var_2925_pad_type_0, strides = var_2925_strides_0, weight = layers_9_self_attn_k_proj_weight_to_fp16, x = input_81_cast_fp16)[name = string("op_2925_cast_fp16")]; string var_2932_pad_type_0 = const()[name = string("op_2932_pad_type_0"), val = string("valid")]; tensor var_2932_strides_0 = const()[name = string("op_2932_strides_0"), val = tensor([1, 1])]; tensor var_2932_pad_0 = const()[name = string("op_2932_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_2932_dilations_0 = const()[name = string("op_2932_dilations_0"), val = tensor([1, 1])]; int32 var_2932_groups_0 = const()[name = string("op_2932_groups_0"), val = int32(1)]; tensor layers_9_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_9_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(255013568)))]; tensor layers_9_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_9_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(257110784)))]; tensor var_2932_cast_fp16 = conv(bias = layers_9_self_attn_v_proj_bias_to_fp16, dilations = var_2932_dilations_0, groups = var_2932_groups_0, pad = var_2932_pad_0, pad_type = var_2932_pad_type_0, strides = var_2932_strides_0, weight = layers_9_self_attn_v_proj_weight_to_fp16, x = input_81_cast_fp16)[name = string("op_2932_cast_fp16")]; tensor tile_27 = const()[name = string("tile_27"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(257112896)))]; int32 var_2933_axis_0 = const()[name = string("op_2933_axis_0"), val = int32(1)]; tensor var_2933_cast_fp16_0, tensor var_2933_cast_fp16_1, tensor var_2933_cast_fp16_2, tensor var_2933_cast_fp16_3, tensor var_2933_cast_fp16_4, tensor var_2933_cast_fp16_5, tensor var_2933_cast_fp16_6, tensor var_2933_cast_fp16_7, tensor var_2933_cast_fp16_8, tensor var_2933_cast_fp16_9, tensor var_2933_cast_fp16_10, tensor var_2933_cast_fp16_11, tensor var_2933_cast_fp16_12, tensor var_2933_cast_fp16_13, tensor var_2933_cast_fp16_14, tensor var_2933_cast_fp16_15 = split(axis = var_2933_axis_0, split_sizes = tile_27, x = var_2918_cast_fp16)[name = string("op_2933_cast_fp16")]; tensor tile_28 = const()[name = string("tile_28"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(257113024)))]; int32 var_2950_axis_0 = const()[name = string("op_2950_axis_0"), val = int32(1)]; tensor var_2950_cast_fp16_0, tensor var_2950_cast_fp16_1, tensor var_2950_cast_fp16_2, tensor var_2950_cast_fp16_3, tensor var_2950_cast_fp16_4, tensor var_2950_cast_fp16_5, tensor var_2950_cast_fp16_6, tensor var_2950_cast_fp16_7, tensor var_2950_cast_fp16_8, tensor var_2950_cast_fp16_9, tensor var_2950_cast_fp16_10, tensor var_2950_cast_fp16_11, tensor var_2950_cast_fp16_12, tensor var_2950_cast_fp16_13, tensor var_2950_cast_fp16_14, tensor var_2950_cast_fp16_15 = split(axis = var_2950_axis_0, split_sizes = tile_28, x = var_2925_cast_fp16)[name = string("op_2950_cast_fp16")]; tensor tile_29 = const()[name = string("tile_29"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(257113152)))]; int32 var_2967_axis_0 = const()[name = string("op_2967_axis_0"), val = int32(1)]; tensor var_2967_cast_fp16_0, tensor var_2967_cast_fp16_1, tensor var_2967_cast_fp16_2, tensor var_2967_cast_fp16_3, tensor var_2967_cast_fp16_4, tensor var_2967_cast_fp16_5, tensor var_2967_cast_fp16_6, tensor var_2967_cast_fp16_7, tensor var_2967_cast_fp16_8, tensor var_2967_cast_fp16_9, tensor var_2967_cast_fp16_10, tensor var_2967_cast_fp16_11, tensor var_2967_cast_fp16_12, tensor var_2967_cast_fp16_13, tensor var_2967_cast_fp16_14, tensor var_2967_cast_fp16_15 = split(axis = var_2967_axis_0, split_sizes = tile_29, x = var_2932_cast_fp16)[name = string("op_2967_cast_fp16")]; tensor transpose_288_perm_0 = const()[name = string("transpose_288_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1444 = const()[name = string("concat_1444"), val = tensor([1, 104, 64])]; tensor transpose_288_cast_fp16 = transpose(perm = transpose_288_perm_0, x = var_2933_cast_fp16_0)[name = string("transpose_3791")]; tensor reshape_432_cast_fp16 = reshape(shape = concat_1444, x = transpose_288_cast_fp16)[name = string("reshape_432_cast_fp16")]; tensor transpose_289_perm_0 = const()[name = string("transpose_289_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1445 = const()[name = string("concat_1445"), val = tensor([1, 64, 104])]; tensor transpose_289_cast_fp16 = transpose(perm = transpose_289_perm_0, x = var_2950_cast_fp16_0)[name = string("transpose_3790")]; tensor reshape_433_cast_fp16 = reshape(shape = concat_1445, x = transpose_289_cast_fp16)[name = string("reshape_433_cast_fp16")]; bool matmul_144_transpose_x_0 = const()[name = string("matmul_144_transpose_x_0"), val = bool(false)]; bool matmul_144_transpose_y_0 = const()[name = string("matmul_144_transpose_y_0"), val = bool(false)]; tensor matmul_144_cast_fp16 = matmul(transpose_x = matmul_144_transpose_x_0, transpose_y = matmul_144_transpose_y_0, x = reshape_432_cast_fp16, y = reshape_433_cast_fp16)[name = string("matmul_144_cast_fp16")]; tensor concat_1449 = const()[name = string("concat_1449"), val = tensor([1, 1, 104, 104])]; tensor reshape_434_cast_fp16 = reshape(shape = concat_1449, x = matmul_144_cast_fp16)[name = string("reshape_434_cast_fp16")]; tensor transpose_2832_perm_0 = const()[name = string("transpose_2832_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2832 = transpose(perm = transpose_2832_perm_0, x = reshape_434_cast_fp16)[name = string("transpose_3789")]; tensor w_579_cast_fp16 = add(x = transpose_2832, y = transpose_2305)[name = string("w_579_cast_fp16")]; tensor var_2989_cast_fp16 = softmax(axis = var_2877, x = w_579_cast_fp16)[name = string("op_2989_cast_fp16")]; string var_2991_equation_0 = const()[name = string("op_2991_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2991_cast_fp16 = einsum(equation = var_2991_equation_0, values = (var_2967_cast_fp16_0, var_2989_cast_fp16))[name = string("op_2991_cast_fp16")]; tensor transpose_290_perm_0 = const()[name = string("transpose_290_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1454 = const()[name = string("concat_1454"), val = tensor([1, 104, 64])]; tensor transpose_290_cast_fp16 = transpose(perm = transpose_290_perm_0, x = var_2933_cast_fp16_1)[name = string("transpose_3788")]; tensor reshape_435_cast_fp16 = reshape(shape = concat_1454, x = transpose_290_cast_fp16)[name = string("reshape_435_cast_fp16")]; tensor transpose_291_perm_0 = const()[name = string("transpose_291_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1455 = const()[name = string("concat_1455"), val = tensor([1, 64, 104])]; tensor transpose_291_cast_fp16 = transpose(perm = transpose_291_perm_0, x = var_2950_cast_fp16_1)[name = string("transpose_3787")]; tensor reshape_436_cast_fp16 = reshape(shape = concat_1455, x = transpose_291_cast_fp16)[name = string("reshape_436_cast_fp16")]; bool matmul_145_transpose_x_0 = const()[name = string("matmul_145_transpose_x_0"), val = bool(false)]; bool matmul_145_transpose_y_0 = const()[name = string("matmul_145_transpose_y_0"), val = bool(false)]; tensor matmul_145_cast_fp16 = matmul(transpose_x = matmul_145_transpose_x_0, transpose_y = matmul_145_transpose_y_0, x = reshape_435_cast_fp16, y = reshape_436_cast_fp16)[name = string("matmul_145_cast_fp16")]; tensor concat_1459 = const()[name = string("concat_1459"), val = tensor([1, 1, 104, 104])]; tensor reshape_437_cast_fp16 = reshape(shape = concat_1459, x = matmul_145_cast_fp16)[name = string("reshape_437_cast_fp16")]; tensor transpose_2833_perm_0 = const()[name = string("transpose_2833_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2833 = transpose(perm = transpose_2833_perm_0, x = reshape_437_cast_fp16)[name = string("transpose_3786")]; tensor w_583_cast_fp16 = add(x = transpose_2833, y = transpose_2305)[name = string("w_583_cast_fp16")]; tensor var_2997_cast_fp16 = softmax(axis = var_2877, x = w_583_cast_fp16)[name = string("op_2997_cast_fp16")]; string var_2999_equation_0 = const()[name = string("op_2999_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_2999_cast_fp16 = einsum(equation = var_2999_equation_0, values = (var_2967_cast_fp16_1, var_2997_cast_fp16))[name = string("op_2999_cast_fp16")]; tensor transpose_292_perm_0 = const()[name = string("transpose_292_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1464 = const()[name = string("concat_1464"), val = tensor([1, 104, 64])]; tensor transpose_292_cast_fp16 = transpose(perm = transpose_292_perm_0, x = var_2933_cast_fp16_2)[name = string("transpose_3785")]; tensor reshape_438_cast_fp16 = reshape(shape = concat_1464, x = transpose_292_cast_fp16)[name = string("reshape_438_cast_fp16")]; tensor transpose_293_perm_0 = const()[name = string("transpose_293_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1465 = const()[name = string("concat_1465"), val = tensor([1, 64, 104])]; tensor transpose_293_cast_fp16 = transpose(perm = transpose_293_perm_0, x = var_2950_cast_fp16_2)[name = string("transpose_3784")]; tensor reshape_439_cast_fp16 = reshape(shape = concat_1465, x = transpose_293_cast_fp16)[name = string("reshape_439_cast_fp16")]; bool matmul_146_transpose_x_0 = const()[name = string("matmul_146_transpose_x_0"), val = bool(false)]; bool matmul_146_transpose_y_0 = const()[name = string("matmul_146_transpose_y_0"), val = bool(false)]; tensor matmul_146_cast_fp16 = matmul(transpose_x = matmul_146_transpose_x_0, transpose_y = matmul_146_transpose_y_0, x = reshape_438_cast_fp16, y = reshape_439_cast_fp16)[name = string("matmul_146_cast_fp16")]; tensor concat_1469 = const()[name = string("concat_1469"), val = tensor([1, 1, 104, 104])]; tensor reshape_440_cast_fp16 = reshape(shape = concat_1469, x = matmul_146_cast_fp16)[name = string("reshape_440_cast_fp16")]; tensor transpose_2834_perm_0 = const()[name = string("transpose_2834_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2834 = transpose(perm = transpose_2834_perm_0, x = reshape_440_cast_fp16)[name = string("transpose_3783")]; tensor w_587_cast_fp16 = add(x = transpose_2834, y = transpose_2305)[name = string("w_587_cast_fp16")]; tensor var_3005_cast_fp16 = softmax(axis = var_2877, x = w_587_cast_fp16)[name = string("op_3005_cast_fp16")]; string var_3007_equation_0 = const()[name = string("op_3007_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3007_cast_fp16 = einsum(equation = var_3007_equation_0, values = (var_2967_cast_fp16_2, var_3005_cast_fp16))[name = string("op_3007_cast_fp16")]; tensor transpose_294_perm_0 = const()[name = string("transpose_294_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1474 = const()[name = string("concat_1474"), val = tensor([1, 104, 64])]; tensor transpose_294_cast_fp16 = transpose(perm = transpose_294_perm_0, x = var_2933_cast_fp16_3)[name = string("transpose_3782")]; tensor reshape_441_cast_fp16 = reshape(shape = concat_1474, x = transpose_294_cast_fp16)[name = string("reshape_441_cast_fp16")]; tensor transpose_295_perm_0 = const()[name = string("transpose_295_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1475 = const()[name = string("concat_1475"), val = tensor([1, 64, 104])]; tensor transpose_295_cast_fp16 = transpose(perm = transpose_295_perm_0, x = var_2950_cast_fp16_3)[name = string("transpose_3781")]; tensor reshape_442_cast_fp16 = reshape(shape = concat_1475, x = transpose_295_cast_fp16)[name = string("reshape_442_cast_fp16")]; bool matmul_147_transpose_x_0 = const()[name = string("matmul_147_transpose_x_0"), val = bool(false)]; bool matmul_147_transpose_y_0 = const()[name = string("matmul_147_transpose_y_0"), val = bool(false)]; tensor matmul_147_cast_fp16 = matmul(transpose_x = matmul_147_transpose_x_0, transpose_y = matmul_147_transpose_y_0, x = reshape_441_cast_fp16, y = reshape_442_cast_fp16)[name = string("matmul_147_cast_fp16")]; tensor concat_1479 = const()[name = string("concat_1479"), val = tensor([1, 1, 104, 104])]; tensor reshape_443_cast_fp16 = reshape(shape = concat_1479, x = matmul_147_cast_fp16)[name = string("reshape_443_cast_fp16")]; tensor transpose_2835_perm_0 = const()[name = string("transpose_2835_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2835 = transpose(perm = transpose_2835_perm_0, x = reshape_443_cast_fp16)[name = string("transpose_3780")]; tensor w_591_cast_fp16 = add(x = transpose_2835, y = transpose_2305)[name = string("w_591_cast_fp16")]; tensor var_3013_cast_fp16 = softmax(axis = var_2877, x = w_591_cast_fp16)[name = string("op_3013_cast_fp16")]; string var_3015_equation_0 = const()[name = string("op_3015_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3015_cast_fp16 = einsum(equation = var_3015_equation_0, values = (var_2967_cast_fp16_3, var_3013_cast_fp16))[name = string("op_3015_cast_fp16")]; tensor transpose_296_perm_0 = const()[name = string("transpose_296_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1484 = const()[name = string("concat_1484"), val = tensor([1, 104, 64])]; tensor transpose_296_cast_fp16 = transpose(perm = transpose_296_perm_0, x = var_2933_cast_fp16_4)[name = string("transpose_3779")]; tensor reshape_444_cast_fp16 = reshape(shape = concat_1484, x = transpose_296_cast_fp16)[name = string("reshape_444_cast_fp16")]; tensor transpose_297_perm_0 = const()[name = string("transpose_297_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1485 = const()[name = string("concat_1485"), val = tensor([1, 64, 104])]; tensor transpose_297_cast_fp16 = transpose(perm = transpose_297_perm_0, x = var_2950_cast_fp16_4)[name = string("transpose_3778")]; tensor reshape_445_cast_fp16 = reshape(shape = concat_1485, x = transpose_297_cast_fp16)[name = string("reshape_445_cast_fp16")]; bool matmul_148_transpose_x_0 = const()[name = string("matmul_148_transpose_x_0"), val = bool(false)]; bool matmul_148_transpose_y_0 = const()[name = string("matmul_148_transpose_y_0"), val = bool(false)]; tensor matmul_148_cast_fp16 = matmul(transpose_x = matmul_148_transpose_x_0, transpose_y = matmul_148_transpose_y_0, x = reshape_444_cast_fp16, y = reshape_445_cast_fp16)[name = string("matmul_148_cast_fp16")]; tensor concat_1489 = const()[name = string("concat_1489"), val = tensor([1, 1, 104, 104])]; tensor reshape_446_cast_fp16 = reshape(shape = concat_1489, x = matmul_148_cast_fp16)[name = string("reshape_446_cast_fp16")]; tensor transpose_2836_perm_0 = const()[name = string("transpose_2836_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2836 = transpose(perm = transpose_2836_perm_0, x = reshape_446_cast_fp16)[name = string("transpose_3777")]; tensor w_595_cast_fp16 = add(x = transpose_2836, y = transpose_2305)[name = string("w_595_cast_fp16")]; tensor var_3021_cast_fp16 = softmax(axis = var_2877, x = w_595_cast_fp16)[name = string("op_3021_cast_fp16")]; string var_3023_equation_0 = const()[name = string("op_3023_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3023_cast_fp16 = einsum(equation = var_3023_equation_0, values = (var_2967_cast_fp16_4, var_3021_cast_fp16))[name = string("op_3023_cast_fp16")]; tensor transpose_298_perm_0 = const()[name = string("transpose_298_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1494 = const()[name = string("concat_1494"), val = tensor([1, 104, 64])]; tensor transpose_298_cast_fp16 = transpose(perm = transpose_298_perm_0, x = var_2933_cast_fp16_5)[name = string("transpose_3776")]; tensor reshape_447_cast_fp16 = reshape(shape = concat_1494, x = transpose_298_cast_fp16)[name = string("reshape_447_cast_fp16")]; tensor transpose_299_perm_0 = const()[name = string("transpose_299_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1495 = const()[name = string("concat_1495"), val = tensor([1, 64, 104])]; tensor transpose_299_cast_fp16 = transpose(perm = transpose_299_perm_0, x = var_2950_cast_fp16_5)[name = string("transpose_3775")]; tensor reshape_448_cast_fp16 = reshape(shape = concat_1495, x = transpose_299_cast_fp16)[name = string("reshape_448_cast_fp16")]; bool matmul_149_transpose_x_0 = const()[name = string("matmul_149_transpose_x_0"), val = bool(false)]; bool matmul_149_transpose_y_0 = const()[name = string("matmul_149_transpose_y_0"), val = bool(false)]; tensor matmul_149_cast_fp16 = matmul(transpose_x = matmul_149_transpose_x_0, transpose_y = matmul_149_transpose_y_0, x = reshape_447_cast_fp16, y = reshape_448_cast_fp16)[name = string("matmul_149_cast_fp16")]; tensor concat_1499 = const()[name = string("concat_1499"), val = tensor([1, 1, 104, 104])]; tensor reshape_449_cast_fp16 = reshape(shape = concat_1499, x = matmul_149_cast_fp16)[name = string("reshape_449_cast_fp16")]; tensor transpose_2837_perm_0 = const()[name = string("transpose_2837_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2837 = transpose(perm = transpose_2837_perm_0, x = reshape_449_cast_fp16)[name = string("transpose_3774")]; tensor w_599_cast_fp16 = add(x = transpose_2837, y = transpose_2305)[name = string("w_599_cast_fp16")]; tensor var_3029_cast_fp16 = softmax(axis = var_2877, x = w_599_cast_fp16)[name = string("op_3029_cast_fp16")]; string var_3031_equation_0 = const()[name = string("op_3031_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3031_cast_fp16 = einsum(equation = var_3031_equation_0, values = (var_2967_cast_fp16_5, var_3029_cast_fp16))[name = string("op_3031_cast_fp16")]; tensor transpose_300_perm_0 = const()[name = string("transpose_300_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1504 = const()[name = string("concat_1504"), val = tensor([1, 104, 64])]; tensor transpose_300_cast_fp16 = transpose(perm = transpose_300_perm_0, x = var_2933_cast_fp16_6)[name = string("transpose_3773")]; tensor reshape_450_cast_fp16 = reshape(shape = concat_1504, x = transpose_300_cast_fp16)[name = string("reshape_450_cast_fp16")]; tensor transpose_301_perm_0 = const()[name = string("transpose_301_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1505 = const()[name = string("concat_1505"), val = tensor([1, 64, 104])]; tensor transpose_301_cast_fp16 = transpose(perm = transpose_301_perm_0, x = var_2950_cast_fp16_6)[name = string("transpose_3772")]; tensor reshape_451_cast_fp16 = reshape(shape = concat_1505, x = transpose_301_cast_fp16)[name = string("reshape_451_cast_fp16")]; bool matmul_150_transpose_x_0 = const()[name = string("matmul_150_transpose_x_0"), val = bool(false)]; bool matmul_150_transpose_y_0 = const()[name = string("matmul_150_transpose_y_0"), val = bool(false)]; tensor matmul_150_cast_fp16 = matmul(transpose_x = matmul_150_transpose_x_0, transpose_y = matmul_150_transpose_y_0, x = reshape_450_cast_fp16, y = reshape_451_cast_fp16)[name = string("matmul_150_cast_fp16")]; tensor concat_1509 = const()[name = string("concat_1509"), val = tensor([1, 1, 104, 104])]; tensor reshape_452_cast_fp16 = reshape(shape = concat_1509, x = matmul_150_cast_fp16)[name = string("reshape_452_cast_fp16")]; tensor transpose_2838_perm_0 = const()[name = string("transpose_2838_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2838 = transpose(perm = transpose_2838_perm_0, x = reshape_452_cast_fp16)[name = string("transpose_3771")]; tensor w_603_cast_fp16 = add(x = transpose_2838, y = transpose_2305)[name = string("w_603_cast_fp16")]; tensor var_3037_cast_fp16 = softmax(axis = var_2877, x = w_603_cast_fp16)[name = string("op_3037_cast_fp16")]; string var_3039_equation_0 = const()[name = string("op_3039_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3039_cast_fp16 = einsum(equation = var_3039_equation_0, values = (var_2967_cast_fp16_6, var_3037_cast_fp16))[name = string("op_3039_cast_fp16")]; tensor transpose_302_perm_0 = const()[name = string("transpose_302_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1514 = const()[name = string("concat_1514"), val = tensor([1, 104, 64])]; tensor transpose_302_cast_fp16 = transpose(perm = transpose_302_perm_0, x = var_2933_cast_fp16_7)[name = string("transpose_3770")]; tensor reshape_453_cast_fp16 = reshape(shape = concat_1514, x = transpose_302_cast_fp16)[name = string("reshape_453_cast_fp16")]; tensor transpose_303_perm_0 = const()[name = string("transpose_303_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1515 = const()[name = string("concat_1515"), val = tensor([1, 64, 104])]; tensor transpose_303_cast_fp16 = transpose(perm = transpose_303_perm_0, x = var_2950_cast_fp16_7)[name = string("transpose_3769")]; tensor reshape_454_cast_fp16 = reshape(shape = concat_1515, x = transpose_303_cast_fp16)[name = string("reshape_454_cast_fp16")]; bool matmul_151_transpose_x_0 = const()[name = string("matmul_151_transpose_x_0"), val = bool(false)]; bool matmul_151_transpose_y_0 = const()[name = string("matmul_151_transpose_y_0"), val = bool(false)]; tensor matmul_151_cast_fp16 = matmul(transpose_x = matmul_151_transpose_x_0, transpose_y = matmul_151_transpose_y_0, x = reshape_453_cast_fp16, y = reshape_454_cast_fp16)[name = string("matmul_151_cast_fp16")]; tensor concat_1519 = const()[name = string("concat_1519"), val = tensor([1, 1, 104, 104])]; tensor reshape_455_cast_fp16 = reshape(shape = concat_1519, x = matmul_151_cast_fp16)[name = string("reshape_455_cast_fp16")]; tensor transpose_2839_perm_0 = const()[name = string("transpose_2839_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2839 = transpose(perm = transpose_2839_perm_0, x = reshape_455_cast_fp16)[name = string("transpose_3768")]; tensor w_607_cast_fp16 = add(x = transpose_2839, y = transpose_2305)[name = string("w_607_cast_fp16")]; tensor var_3045_cast_fp16 = softmax(axis = var_2877, x = w_607_cast_fp16)[name = string("op_3045_cast_fp16")]; string var_3047_equation_0 = const()[name = string("op_3047_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3047_cast_fp16 = einsum(equation = var_3047_equation_0, values = (var_2967_cast_fp16_7, var_3045_cast_fp16))[name = string("op_3047_cast_fp16")]; tensor transpose_304_perm_0 = const()[name = string("transpose_304_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1524 = const()[name = string("concat_1524"), val = tensor([1, 104, 64])]; tensor transpose_304_cast_fp16 = transpose(perm = transpose_304_perm_0, x = var_2933_cast_fp16_8)[name = string("transpose_3767")]; tensor reshape_456_cast_fp16 = reshape(shape = concat_1524, x = transpose_304_cast_fp16)[name = string("reshape_456_cast_fp16")]; tensor transpose_305_perm_0 = const()[name = string("transpose_305_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1525 = const()[name = string("concat_1525"), val = tensor([1, 64, 104])]; tensor transpose_305_cast_fp16 = transpose(perm = transpose_305_perm_0, x = var_2950_cast_fp16_8)[name = string("transpose_3766")]; tensor reshape_457_cast_fp16 = reshape(shape = concat_1525, x = transpose_305_cast_fp16)[name = string("reshape_457_cast_fp16")]; bool matmul_152_transpose_x_0 = const()[name = string("matmul_152_transpose_x_0"), val = bool(false)]; bool matmul_152_transpose_y_0 = const()[name = string("matmul_152_transpose_y_0"), val = bool(false)]; tensor matmul_152_cast_fp16 = matmul(transpose_x = matmul_152_transpose_x_0, transpose_y = matmul_152_transpose_y_0, x = reshape_456_cast_fp16, y = reshape_457_cast_fp16)[name = string("matmul_152_cast_fp16")]; tensor concat_1529 = const()[name = string("concat_1529"), val = tensor([1, 1, 104, 104])]; tensor reshape_458_cast_fp16 = reshape(shape = concat_1529, x = matmul_152_cast_fp16)[name = string("reshape_458_cast_fp16")]; tensor transpose_2840_perm_0 = const()[name = string("transpose_2840_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2840 = transpose(perm = transpose_2840_perm_0, x = reshape_458_cast_fp16)[name = string("transpose_3765")]; tensor w_611_cast_fp16 = add(x = transpose_2840, y = transpose_2305)[name = string("w_611_cast_fp16")]; tensor var_3053_cast_fp16 = softmax(axis = var_2877, x = w_611_cast_fp16)[name = string("op_3053_cast_fp16")]; string var_3055_equation_0 = const()[name = string("op_3055_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3055_cast_fp16 = einsum(equation = var_3055_equation_0, values = (var_2967_cast_fp16_8, var_3053_cast_fp16))[name = string("op_3055_cast_fp16")]; tensor transpose_306_perm_0 = const()[name = string("transpose_306_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1534 = const()[name = string("concat_1534"), val = tensor([1, 104, 64])]; tensor transpose_306_cast_fp16 = transpose(perm = transpose_306_perm_0, x = var_2933_cast_fp16_9)[name = string("transpose_3764")]; tensor reshape_459_cast_fp16 = reshape(shape = concat_1534, x = transpose_306_cast_fp16)[name = string("reshape_459_cast_fp16")]; tensor transpose_307_perm_0 = const()[name = string("transpose_307_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1535 = const()[name = string("concat_1535"), val = tensor([1, 64, 104])]; tensor transpose_307_cast_fp16 = transpose(perm = transpose_307_perm_0, x = var_2950_cast_fp16_9)[name = string("transpose_3763")]; tensor reshape_460_cast_fp16 = reshape(shape = concat_1535, x = transpose_307_cast_fp16)[name = string("reshape_460_cast_fp16")]; bool matmul_153_transpose_x_0 = const()[name = string("matmul_153_transpose_x_0"), val = bool(false)]; bool matmul_153_transpose_y_0 = const()[name = string("matmul_153_transpose_y_0"), val = bool(false)]; tensor matmul_153_cast_fp16 = matmul(transpose_x = matmul_153_transpose_x_0, transpose_y = matmul_153_transpose_y_0, x = reshape_459_cast_fp16, y = reshape_460_cast_fp16)[name = string("matmul_153_cast_fp16")]; tensor concat_1539 = const()[name = string("concat_1539"), val = tensor([1, 1, 104, 104])]; tensor reshape_461_cast_fp16 = reshape(shape = concat_1539, x = matmul_153_cast_fp16)[name = string("reshape_461_cast_fp16")]; tensor transpose_2841_perm_0 = const()[name = string("transpose_2841_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2841 = transpose(perm = transpose_2841_perm_0, x = reshape_461_cast_fp16)[name = string("transpose_3762")]; tensor w_615_cast_fp16 = add(x = transpose_2841, y = transpose_2305)[name = string("w_615_cast_fp16")]; tensor var_3061_cast_fp16 = softmax(axis = var_2877, x = w_615_cast_fp16)[name = string("op_3061_cast_fp16")]; string var_3063_equation_0 = const()[name = string("op_3063_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3063_cast_fp16 = einsum(equation = var_3063_equation_0, values = (var_2967_cast_fp16_9, var_3061_cast_fp16))[name = string("op_3063_cast_fp16")]; tensor transpose_308_perm_0 = const()[name = string("transpose_308_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1544 = const()[name = string("concat_1544"), val = tensor([1, 104, 64])]; tensor transpose_308_cast_fp16 = transpose(perm = transpose_308_perm_0, x = var_2933_cast_fp16_10)[name = string("transpose_3761")]; tensor reshape_462_cast_fp16 = reshape(shape = concat_1544, x = transpose_308_cast_fp16)[name = string("reshape_462_cast_fp16")]; tensor transpose_309_perm_0 = const()[name = string("transpose_309_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1545 = const()[name = string("concat_1545"), val = tensor([1, 64, 104])]; tensor transpose_309_cast_fp16 = transpose(perm = transpose_309_perm_0, x = var_2950_cast_fp16_10)[name = string("transpose_3760")]; tensor reshape_463_cast_fp16 = reshape(shape = concat_1545, x = transpose_309_cast_fp16)[name = string("reshape_463_cast_fp16")]; bool matmul_154_transpose_x_0 = const()[name = string("matmul_154_transpose_x_0"), val = bool(false)]; bool matmul_154_transpose_y_0 = const()[name = string("matmul_154_transpose_y_0"), val = bool(false)]; tensor matmul_154_cast_fp16 = matmul(transpose_x = matmul_154_transpose_x_0, transpose_y = matmul_154_transpose_y_0, x = reshape_462_cast_fp16, y = reshape_463_cast_fp16)[name = string("matmul_154_cast_fp16")]; tensor concat_1549 = const()[name = string("concat_1549"), val = tensor([1, 1, 104, 104])]; tensor reshape_464_cast_fp16 = reshape(shape = concat_1549, x = matmul_154_cast_fp16)[name = string("reshape_464_cast_fp16")]; tensor transpose_2842_perm_0 = const()[name = string("transpose_2842_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2842 = transpose(perm = transpose_2842_perm_0, x = reshape_464_cast_fp16)[name = string("transpose_3759")]; tensor w_619_cast_fp16 = add(x = transpose_2842, y = transpose_2305)[name = string("w_619_cast_fp16")]; tensor var_3069_cast_fp16 = softmax(axis = var_2877, x = w_619_cast_fp16)[name = string("op_3069_cast_fp16")]; string var_3071_equation_0 = const()[name = string("op_3071_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3071_cast_fp16 = einsum(equation = var_3071_equation_0, values = (var_2967_cast_fp16_10, var_3069_cast_fp16))[name = string("op_3071_cast_fp16")]; tensor transpose_310_perm_0 = const()[name = string("transpose_310_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1554 = const()[name = string("concat_1554"), val = tensor([1, 104, 64])]; tensor transpose_310_cast_fp16 = transpose(perm = transpose_310_perm_0, x = var_2933_cast_fp16_11)[name = string("transpose_3758")]; tensor reshape_465_cast_fp16 = reshape(shape = concat_1554, x = transpose_310_cast_fp16)[name = string("reshape_465_cast_fp16")]; tensor transpose_311_perm_0 = const()[name = string("transpose_311_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1555 = const()[name = string("concat_1555"), val = tensor([1, 64, 104])]; tensor transpose_311_cast_fp16 = transpose(perm = transpose_311_perm_0, x = var_2950_cast_fp16_11)[name = string("transpose_3757")]; tensor reshape_466_cast_fp16 = reshape(shape = concat_1555, x = transpose_311_cast_fp16)[name = string("reshape_466_cast_fp16")]; bool matmul_155_transpose_x_0 = const()[name = string("matmul_155_transpose_x_0"), val = bool(false)]; bool matmul_155_transpose_y_0 = const()[name = string("matmul_155_transpose_y_0"), val = bool(false)]; tensor matmul_155_cast_fp16 = matmul(transpose_x = matmul_155_transpose_x_0, transpose_y = matmul_155_transpose_y_0, x = reshape_465_cast_fp16, y = reshape_466_cast_fp16)[name = string("matmul_155_cast_fp16")]; tensor concat_1559 = const()[name = string("concat_1559"), val = tensor([1, 1, 104, 104])]; tensor reshape_467_cast_fp16 = reshape(shape = concat_1559, x = matmul_155_cast_fp16)[name = string("reshape_467_cast_fp16")]; tensor transpose_2843_perm_0 = const()[name = string("transpose_2843_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2843 = transpose(perm = transpose_2843_perm_0, x = reshape_467_cast_fp16)[name = string("transpose_3756")]; tensor w_623_cast_fp16 = add(x = transpose_2843, y = transpose_2305)[name = string("w_623_cast_fp16")]; tensor var_3077_cast_fp16 = softmax(axis = var_2877, x = w_623_cast_fp16)[name = string("op_3077_cast_fp16")]; string var_3079_equation_0 = const()[name = string("op_3079_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3079_cast_fp16 = einsum(equation = var_3079_equation_0, values = (var_2967_cast_fp16_11, var_3077_cast_fp16))[name = string("op_3079_cast_fp16")]; tensor transpose_312_perm_0 = const()[name = string("transpose_312_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1564 = const()[name = string("concat_1564"), val = tensor([1, 104, 64])]; tensor transpose_312_cast_fp16 = transpose(perm = transpose_312_perm_0, x = var_2933_cast_fp16_12)[name = string("transpose_3755")]; tensor reshape_468_cast_fp16 = reshape(shape = concat_1564, x = transpose_312_cast_fp16)[name = string("reshape_468_cast_fp16")]; tensor transpose_313_perm_0 = const()[name = string("transpose_313_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1565 = const()[name = string("concat_1565"), val = tensor([1, 64, 104])]; tensor transpose_313_cast_fp16 = transpose(perm = transpose_313_perm_0, x = var_2950_cast_fp16_12)[name = string("transpose_3754")]; tensor reshape_469_cast_fp16 = reshape(shape = concat_1565, x = transpose_313_cast_fp16)[name = string("reshape_469_cast_fp16")]; bool matmul_156_transpose_x_0 = const()[name = string("matmul_156_transpose_x_0"), val = bool(false)]; bool matmul_156_transpose_y_0 = const()[name = string("matmul_156_transpose_y_0"), val = bool(false)]; tensor matmul_156_cast_fp16 = matmul(transpose_x = matmul_156_transpose_x_0, transpose_y = matmul_156_transpose_y_0, x = reshape_468_cast_fp16, y = reshape_469_cast_fp16)[name = string("matmul_156_cast_fp16")]; tensor concat_1569 = const()[name = string("concat_1569"), val = tensor([1, 1, 104, 104])]; tensor reshape_470_cast_fp16 = reshape(shape = concat_1569, x = matmul_156_cast_fp16)[name = string("reshape_470_cast_fp16")]; tensor transpose_2844_perm_0 = const()[name = string("transpose_2844_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2844 = transpose(perm = transpose_2844_perm_0, x = reshape_470_cast_fp16)[name = string("transpose_3753")]; tensor w_627_cast_fp16 = add(x = transpose_2844, y = transpose_2305)[name = string("w_627_cast_fp16")]; tensor var_3085_cast_fp16 = softmax(axis = var_2877, x = w_627_cast_fp16)[name = string("op_3085_cast_fp16")]; string var_3087_equation_0 = const()[name = string("op_3087_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3087_cast_fp16 = einsum(equation = var_3087_equation_0, values = (var_2967_cast_fp16_12, var_3085_cast_fp16))[name = string("op_3087_cast_fp16")]; tensor transpose_314_perm_0 = const()[name = string("transpose_314_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1574 = const()[name = string("concat_1574"), val = tensor([1, 104, 64])]; tensor transpose_314_cast_fp16 = transpose(perm = transpose_314_perm_0, x = var_2933_cast_fp16_13)[name = string("transpose_3752")]; tensor reshape_471_cast_fp16 = reshape(shape = concat_1574, x = transpose_314_cast_fp16)[name = string("reshape_471_cast_fp16")]; tensor transpose_315_perm_0 = const()[name = string("transpose_315_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1575 = const()[name = string("concat_1575"), val = tensor([1, 64, 104])]; tensor transpose_315_cast_fp16 = transpose(perm = transpose_315_perm_0, x = var_2950_cast_fp16_13)[name = string("transpose_3751")]; tensor reshape_472_cast_fp16 = reshape(shape = concat_1575, x = transpose_315_cast_fp16)[name = string("reshape_472_cast_fp16")]; bool matmul_157_transpose_x_0 = const()[name = string("matmul_157_transpose_x_0"), val = bool(false)]; bool matmul_157_transpose_y_0 = const()[name = string("matmul_157_transpose_y_0"), val = bool(false)]; tensor matmul_157_cast_fp16 = matmul(transpose_x = matmul_157_transpose_x_0, transpose_y = matmul_157_transpose_y_0, x = reshape_471_cast_fp16, y = reshape_472_cast_fp16)[name = string("matmul_157_cast_fp16")]; tensor concat_1579 = const()[name = string("concat_1579"), val = tensor([1, 1, 104, 104])]; tensor reshape_473_cast_fp16 = reshape(shape = concat_1579, x = matmul_157_cast_fp16)[name = string("reshape_473_cast_fp16")]; tensor transpose_2845_perm_0 = const()[name = string("transpose_2845_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2845 = transpose(perm = transpose_2845_perm_0, x = reshape_473_cast_fp16)[name = string("transpose_3750")]; tensor w_631_cast_fp16 = add(x = transpose_2845, y = transpose_2305)[name = string("w_631_cast_fp16")]; tensor var_3093_cast_fp16 = softmax(axis = var_2877, x = w_631_cast_fp16)[name = string("op_3093_cast_fp16")]; string var_3095_equation_0 = const()[name = string("op_3095_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3095_cast_fp16 = einsum(equation = var_3095_equation_0, values = (var_2967_cast_fp16_13, var_3093_cast_fp16))[name = string("op_3095_cast_fp16")]; tensor transpose_316_perm_0 = const()[name = string("transpose_316_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1584 = const()[name = string("concat_1584"), val = tensor([1, 104, 64])]; tensor transpose_316_cast_fp16 = transpose(perm = transpose_316_perm_0, x = var_2933_cast_fp16_14)[name = string("transpose_3749")]; tensor reshape_474_cast_fp16 = reshape(shape = concat_1584, x = transpose_316_cast_fp16)[name = string("reshape_474_cast_fp16")]; tensor transpose_317_perm_0 = const()[name = string("transpose_317_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1585 = const()[name = string("concat_1585"), val = tensor([1, 64, 104])]; tensor transpose_317_cast_fp16 = transpose(perm = transpose_317_perm_0, x = var_2950_cast_fp16_14)[name = string("transpose_3748")]; tensor reshape_475_cast_fp16 = reshape(shape = concat_1585, x = transpose_317_cast_fp16)[name = string("reshape_475_cast_fp16")]; bool matmul_158_transpose_x_0 = const()[name = string("matmul_158_transpose_x_0"), val = bool(false)]; bool matmul_158_transpose_y_0 = const()[name = string("matmul_158_transpose_y_0"), val = bool(false)]; tensor matmul_158_cast_fp16 = matmul(transpose_x = matmul_158_transpose_x_0, transpose_y = matmul_158_transpose_y_0, x = reshape_474_cast_fp16, y = reshape_475_cast_fp16)[name = string("matmul_158_cast_fp16")]; tensor concat_1589 = const()[name = string("concat_1589"), val = tensor([1, 1, 104, 104])]; tensor reshape_476_cast_fp16 = reshape(shape = concat_1589, x = matmul_158_cast_fp16)[name = string("reshape_476_cast_fp16")]; tensor transpose_2846_perm_0 = const()[name = string("transpose_2846_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2846 = transpose(perm = transpose_2846_perm_0, x = reshape_476_cast_fp16)[name = string("transpose_3747")]; tensor w_635_cast_fp16 = add(x = transpose_2846, y = transpose_2305)[name = string("w_635_cast_fp16")]; tensor var_3101_cast_fp16 = softmax(axis = var_2877, x = w_635_cast_fp16)[name = string("op_3101_cast_fp16")]; string var_3103_equation_0 = const()[name = string("op_3103_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3103_cast_fp16 = einsum(equation = var_3103_equation_0, values = (var_2967_cast_fp16_14, var_3101_cast_fp16))[name = string("op_3103_cast_fp16")]; tensor transpose_318_perm_0 = const()[name = string("transpose_318_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1594 = const()[name = string("concat_1594"), val = tensor([1, 104, 64])]; tensor transpose_318_cast_fp16 = transpose(perm = transpose_318_perm_0, x = var_2933_cast_fp16_15)[name = string("transpose_3746")]; tensor reshape_477_cast_fp16 = reshape(shape = concat_1594, x = transpose_318_cast_fp16)[name = string("reshape_477_cast_fp16")]; tensor transpose_319_perm_0 = const()[name = string("transpose_319_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1595 = const()[name = string("concat_1595"), val = tensor([1, 64, 104])]; tensor transpose_319_cast_fp16 = transpose(perm = transpose_319_perm_0, x = var_2950_cast_fp16_15)[name = string("transpose_3745")]; tensor reshape_478_cast_fp16 = reshape(shape = concat_1595, x = transpose_319_cast_fp16)[name = string("reshape_478_cast_fp16")]; bool matmul_159_transpose_x_0 = const()[name = string("matmul_159_transpose_x_0"), val = bool(false)]; bool matmul_159_transpose_y_0 = const()[name = string("matmul_159_transpose_y_0"), val = bool(false)]; tensor matmul_159_cast_fp16 = matmul(transpose_x = matmul_159_transpose_x_0, transpose_y = matmul_159_transpose_y_0, x = reshape_477_cast_fp16, y = reshape_478_cast_fp16)[name = string("matmul_159_cast_fp16")]; tensor concat_1599 = const()[name = string("concat_1599"), val = tensor([1, 1, 104, 104])]; tensor reshape_479_cast_fp16 = reshape(shape = concat_1599, x = matmul_159_cast_fp16)[name = string("reshape_479_cast_fp16")]; tensor transpose_2847_perm_0 = const()[name = string("transpose_2847_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2847 = transpose(perm = transpose_2847_perm_0, x = reshape_479_cast_fp16)[name = string("transpose_3744")]; tensor w_639_cast_fp16 = add(x = transpose_2847, y = transpose_2305)[name = string("w_639_cast_fp16")]; tensor var_3109_cast_fp16 = softmax(axis = var_2877, x = w_639_cast_fp16)[name = string("op_3109_cast_fp16")]; string var_3111_equation_0 = const()[name = string("op_3111_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3111_cast_fp16 = einsum(equation = var_3111_equation_0, values = (var_2967_cast_fp16_15, var_3109_cast_fp16))[name = string("op_3111_cast_fp16")]; bool input_83_interleave_0 = const()[name = string("input_83_interleave_0"), val = bool(false)]; tensor input_83_cast_fp16 = concat(axis = var_2877, interleave = input_83_interleave_0, values = (var_2991_cast_fp16, var_2999_cast_fp16, var_3007_cast_fp16, var_3015_cast_fp16, var_3023_cast_fp16, var_3031_cast_fp16, var_3039_cast_fp16, var_3047_cast_fp16, var_3055_cast_fp16, var_3063_cast_fp16, var_3071_cast_fp16, var_3079_cast_fp16, var_3087_cast_fp16, var_3095_cast_fp16, var_3103_cast_fp16, var_3111_cast_fp16))[name = string("input_83_cast_fp16")]; string var_3120_pad_type_0 = const()[name = string("op_3120_pad_type_0"), val = string("valid")]; tensor var_3120_strides_0 = const()[name = string("op_3120_strides_0"), val = tensor([1, 1])]; tensor var_3120_pad_0 = const()[name = string("op_3120_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3120_dilations_0 = const()[name = string("op_3120_dilations_0"), val = tensor([1, 1])]; int32 var_3120_groups_0 = const()[name = string("op_3120_groups_0"), val = int32(1)]; tensor layers_9_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_9_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(257113280)))]; tensor layers_9_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_9_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(259210496)))]; tensor var_3120_cast_fp16 = conv(bias = layers_9_self_attn_out_proj_bias_to_fp16, dilations = var_3120_dilations_0, groups = var_3120_groups_0, pad = var_3120_pad_0, pad_type = var_3120_pad_type_0, strides = var_3120_strides_0, weight = layers_9_self_attn_out_proj_weight_to_fp16, x = input_83_cast_fp16)[name = string("op_3120_cast_fp16")]; tensor x_107_cast_fp16 = add(x = x_103_cast_fp16, y = var_3120_cast_fp16)[name = string("x_107_cast_fp16")]; tensor mu_39_axes_0 = const()[name = string("mu_39_axes_0"), val = tensor([1])]; bool mu_39_keep_dims_0 = const()[name = string("mu_39_keep_dims_0"), val = bool(true)]; tensor mu_39_cast_fp16 = reduce_mean(axes = mu_39_axes_0, keep_dims = mu_39_keep_dims_0, x = x_107_cast_fp16)[name = string("mu_39_cast_fp16")]; tensor var_3126_cast_fp16 = sub(x = x_107_cast_fp16, y = mu_39_cast_fp16)[name = string("op_3126_cast_fp16")]; fp16 var_2880_promoted_1_to_fp16 = const()[name = string("op_2880_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_3127_cast_fp16 = pow(x = var_3126_cast_fp16, y = var_2880_promoted_1_to_fp16)[name = string("op_3127_cast_fp16")]; tensor var_39_axes_0 = const()[name = string("var_39_axes_0"), val = tensor([1])]; bool var_39_keep_dims_0 = const()[name = string("var_39_keep_dims_0"), val = bool(true)]; tensor var_39_cast_fp16 = reduce_mean(axes = var_39_axes_0, keep_dims = var_39_keep_dims_0, x = var_3127_cast_fp16)[name = string("var_39_cast_fp16")]; fp16 var_3131_to_fp16 = const()[name = string("op_3131_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_3132_cast_fp16 = add(x = var_39_cast_fp16, y = var_3131_to_fp16)[name = string("op_3132_cast_fp16")]; fp32 var_3133_epsilon_0 = const()[name = string("op_3133_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_3133_cast_fp16 = rsqrt(epsilon = var_3133_epsilon_0, x = var_3132_cast_fp16)[name = string("op_3133_cast_fp16")]; tensor x_109_cast_fp16 = mul(x = var_3126_cast_fp16, y = var_3133_cast_fp16)[name = string("x_109_cast_fp16")]; tensor input_85_gamma_0_to_fp16 = const()[name = string("input_85_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(259212608)))]; tensor input_85_beta_0_to_fp16 = const()[name = string("input_85_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(259214720)))]; fp16 input_85_epsilon_0_to_fp16 = const()[name = string("input_85_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_85_cast_fp16 = batch_norm(beta = input_85_beta_0_to_fp16, epsilon = input_85_epsilon_0_to_fp16, gamma = input_85_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_109_cast_fp16)[name = string("input_85_cast_fp16")]; string x_111_pad_type_0 = const()[name = string("x_111_pad_type_0"), val = string("valid")]; tensor x_111_strides_0 = const()[name = string("x_111_strides_0"), val = tensor([1, 1])]; tensor x_111_pad_0 = const()[name = string("x_111_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_111_dilations_0 = const()[name = string("x_111_dilations_0"), val = tensor([1, 1])]; int32 x_111_groups_0 = const()[name = string("x_111_groups_0"), val = int32(1)]; tensor layers_9_fc1_weight_to_fp16 = const()[name = string("layers_9_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(259216832)))]; tensor layers_9_fc1_bias_to_fp16 = const()[name = string("layers_9_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(267605504)))]; tensor x_111_cast_fp16 = conv(bias = layers_9_fc1_bias_to_fp16, dilations = x_111_dilations_0, groups = x_111_groups_0, pad = x_111_pad_0, pad_type = x_111_pad_type_0, strides = x_111_strides_0, weight = layers_9_fc1_weight_to_fp16, x = input_85_cast_fp16)[name = string("x_111_cast_fp16")]; fp16 var_3148_to_fp16 = const()[name = string("op_3148_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_3149_cast_fp16 = mul(x = x_111_cast_fp16, y = var_3148_to_fp16)[name = string("op_3149_cast_fp16")]; tensor var_3150_cast_fp16 = mul(x = var_3149_cast_fp16, y = x_111_cast_fp16)[name = string("op_3150_cast_fp16")]; tensor var_3151_cast_fp16 = mul(x = var_3150_cast_fp16, y = x_111_cast_fp16)[name = string("op_3151_cast_fp16")]; tensor var_3152_cast_fp16 = add(x = x_111_cast_fp16, y = var_3151_cast_fp16)[name = string("op_3152_cast_fp16")]; fp16 var_3153_to_fp16 = const()[name = string("op_3153_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_25_cast_fp16 = mul(x = var_3152_cast_fp16, y = var_3153_to_fp16)[name = string("u_25_cast_fp16")]; fp16 var_3155_to_fp16 = const()[name = string("op_3155_to_fp16"), val = fp16(0x1p-1)]; tensor var_3156_cast_fp16 = mul(x = x_111_cast_fp16, y = var_3155_to_fp16)[name = string("op_3156_cast_fp16")]; tensor var_3157_cast_fp16 = tanh(x = u_25_cast_fp16)[name = string("op_3157_cast_fp16")]; fp16 var_3158_to_fp16 = const()[name = string("op_3158_to_fp16"), val = fp16(0x1p+0)]; tensor var_3159_cast_fp16 = add(x = var_3157_cast_fp16, y = var_3158_to_fp16)[name = string("op_3159_cast_fp16")]; tensor input_87_cast_fp16 = mul(x = var_3156_cast_fp16, y = var_3159_cast_fp16)[name = string("input_87_cast_fp16")]; string h_19_pad_type_0 = const()[name = string("h_19_pad_type_0"), val = string("valid")]; tensor h_19_strides_0 = const()[name = string("h_19_strides_0"), val = tensor([1, 1])]; tensor h_19_pad_0 = const()[name = string("h_19_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_19_dilations_0 = const()[name = string("h_19_dilations_0"), val = tensor([1, 1])]; int32 h_19_groups_0 = const()[name = string("h_19_groups_0"), val = int32(1)]; tensor layers_9_fc2_weight_to_fp16 = const()[name = string("layers_9_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(267613760)))]; tensor layers_9_fc2_bias_to_fp16 = const()[name = string("layers_9_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(276002432)))]; tensor h_19_cast_fp16 = conv(bias = layers_9_fc2_bias_to_fp16, dilations = h_19_dilations_0, groups = h_19_groups_0, pad = h_19_pad_0, pad_type = h_19_pad_type_0, strides = h_19_strides_0, weight = layers_9_fc2_weight_to_fp16, x = input_87_cast_fp16)[name = string("h_19_cast_fp16")]; tensor x_113_cast_fp16 = add(x = x_107_cast_fp16, y = h_19_cast_fp16)[name = string("x_113_cast_fp16")]; int32 var_3175 = const()[name = string("op_3175"), val = int32(1)]; tensor mu_41_axes_0 = const()[name = string("mu_41_axes_0"), val = tensor([1])]; bool mu_41_keep_dims_0 = const()[name = string("mu_41_keep_dims_0"), val = bool(true)]; tensor mu_41_cast_fp16 = reduce_mean(axes = mu_41_axes_0, keep_dims = mu_41_keep_dims_0, x = x_113_cast_fp16)[name = string("mu_41_cast_fp16")]; tensor var_3189_cast_fp16 = sub(x = x_113_cast_fp16, y = mu_41_cast_fp16)[name = string("op_3189_cast_fp16")]; fp16 var_3178_promoted_to_fp16 = const()[name = string("op_3178_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_3190_cast_fp16 = pow(x = var_3189_cast_fp16, y = var_3178_promoted_to_fp16)[name = string("op_3190_cast_fp16")]; tensor var_41_axes_0 = const()[name = string("var_41_axes_0"), val = tensor([1])]; bool var_41_keep_dims_0 = const()[name = string("var_41_keep_dims_0"), val = bool(true)]; tensor var_41_cast_fp16 = reduce_mean(axes = var_41_axes_0, keep_dims = var_41_keep_dims_0, x = var_3190_cast_fp16)[name = string("var_41_cast_fp16")]; fp16 var_3194_to_fp16 = const()[name = string("op_3194_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_3195_cast_fp16 = add(x = var_41_cast_fp16, y = var_3194_to_fp16)[name = string("op_3195_cast_fp16")]; fp32 var_3196_epsilon_0 = const()[name = string("op_3196_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_3196_cast_fp16 = rsqrt(epsilon = var_3196_epsilon_0, x = var_3195_cast_fp16)[name = string("op_3196_cast_fp16")]; tensor x_115_cast_fp16 = mul(x = var_3189_cast_fp16, y = var_3196_cast_fp16)[name = string("x_115_cast_fp16")]; tensor input_89_gamma_0_to_fp16 = const()[name = string("input_89_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(276004544)))]; tensor input_89_beta_0_to_fp16 = const()[name = string("input_89_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(276006656)))]; fp16 input_89_epsilon_0_to_fp16 = const()[name = string("input_89_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_89_cast_fp16 = batch_norm(beta = input_89_beta_0_to_fp16, epsilon = input_89_epsilon_0_to_fp16, gamma = input_89_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_115_cast_fp16)[name = string("input_89_cast_fp16")]; string var_3214_pad_type_0 = const()[name = string("op_3214_pad_type_0"), val = string("valid")]; tensor var_3214_strides_0 = const()[name = string("op_3214_strides_0"), val = tensor([1, 1])]; tensor var_3214_pad_0 = const()[name = string("op_3214_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3214_dilations_0 = const()[name = string("op_3214_dilations_0"), val = tensor([1, 1])]; int32 var_3214_groups_0 = const()[name = string("op_3214_groups_0"), val = int32(1)]; tensor var_3216_weight_0_to_fp16 = const()[name = string("op_3216_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(276008768)))]; tensor var_3216_bias_0_to_fp16 = const()[name = string("op_3216_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(278105984)))]; tensor var_3216_cast_fp16 = conv(bias = var_3216_bias_0_to_fp16, dilations = var_3214_dilations_0, groups = var_3214_groups_0, pad = var_3214_pad_0, pad_type = var_3214_pad_type_0, strides = var_3214_strides_0, weight = var_3216_weight_0_to_fp16, x = input_89_cast_fp16)[name = string("op_3216_cast_fp16")]; string var_3223_pad_type_0 = const()[name = string("op_3223_pad_type_0"), val = string("valid")]; tensor var_3223_strides_0 = const()[name = string("op_3223_strides_0"), val = tensor([1, 1])]; tensor var_3223_pad_0 = const()[name = string("op_3223_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3223_dilations_0 = const()[name = string("op_3223_dilations_0"), val = tensor([1, 1])]; int32 var_3223_groups_0 = const()[name = string("op_3223_groups_0"), val = int32(1)]; tensor layers_10_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_10_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(278108096)))]; tensor layers_10_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_10_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(280205312)))]; tensor var_3223_cast_fp16 = conv(bias = layers_10_self_attn_k_proj_bias_to_fp16, dilations = var_3223_dilations_0, groups = var_3223_groups_0, pad = var_3223_pad_0, pad_type = var_3223_pad_type_0, strides = var_3223_strides_0, weight = layers_10_self_attn_k_proj_weight_to_fp16, x = input_89_cast_fp16)[name = string("op_3223_cast_fp16")]; string var_3230_pad_type_0 = const()[name = string("op_3230_pad_type_0"), val = string("valid")]; tensor var_3230_strides_0 = const()[name = string("op_3230_strides_0"), val = tensor([1, 1])]; tensor var_3230_pad_0 = const()[name = string("op_3230_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3230_dilations_0 = const()[name = string("op_3230_dilations_0"), val = tensor([1, 1])]; int32 var_3230_groups_0 = const()[name = string("op_3230_groups_0"), val = int32(1)]; tensor layers_10_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_10_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(280207424)))]; tensor layers_10_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_10_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(282304640)))]; tensor var_3230_cast_fp16 = conv(bias = layers_10_self_attn_v_proj_bias_to_fp16, dilations = var_3230_dilations_0, groups = var_3230_groups_0, pad = var_3230_pad_0, pad_type = var_3230_pad_type_0, strides = var_3230_strides_0, weight = layers_10_self_attn_v_proj_weight_to_fp16, x = input_89_cast_fp16)[name = string("op_3230_cast_fp16")]; tensor tile_30 = const()[name = string("tile_30"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(282306752)))]; int32 var_3231_axis_0 = const()[name = string("op_3231_axis_0"), val = int32(1)]; tensor var_3231_cast_fp16_0, tensor var_3231_cast_fp16_1, tensor var_3231_cast_fp16_2, tensor var_3231_cast_fp16_3, tensor var_3231_cast_fp16_4, tensor var_3231_cast_fp16_5, tensor var_3231_cast_fp16_6, tensor var_3231_cast_fp16_7, tensor var_3231_cast_fp16_8, tensor var_3231_cast_fp16_9, tensor var_3231_cast_fp16_10, tensor var_3231_cast_fp16_11, tensor var_3231_cast_fp16_12, tensor var_3231_cast_fp16_13, tensor var_3231_cast_fp16_14, tensor var_3231_cast_fp16_15 = split(axis = var_3231_axis_0, split_sizes = tile_30, x = var_3216_cast_fp16)[name = string("op_3231_cast_fp16")]; tensor tile_31 = const()[name = string("tile_31"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(282306880)))]; int32 var_3248_axis_0 = const()[name = string("op_3248_axis_0"), val = int32(1)]; tensor var_3248_cast_fp16_0, tensor var_3248_cast_fp16_1, tensor var_3248_cast_fp16_2, tensor var_3248_cast_fp16_3, tensor var_3248_cast_fp16_4, tensor var_3248_cast_fp16_5, tensor var_3248_cast_fp16_6, tensor var_3248_cast_fp16_7, tensor var_3248_cast_fp16_8, tensor var_3248_cast_fp16_9, tensor var_3248_cast_fp16_10, tensor var_3248_cast_fp16_11, tensor var_3248_cast_fp16_12, tensor var_3248_cast_fp16_13, tensor var_3248_cast_fp16_14, tensor var_3248_cast_fp16_15 = split(axis = var_3248_axis_0, split_sizes = tile_31, x = var_3223_cast_fp16)[name = string("op_3248_cast_fp16")]; tensor tile_32 = const()[name = string("tile_32"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(282307008)))]; int32 var_3265_axis_0 = const()[name = string("op_3265_axis_0"), val = int32(1)]; tensor var_3265_cast_fp16_0, tensor var_3265_cast_fp16_1, tensor var_3265_cast_fp16_2, tensor var_3265_cast_fp16_3, tensor var_3265_cast_fp16_4, tensor var_3265_cast_fp16_5, tensor var_3265_cast_fp16_6, tensor var_3265_cast_fp16_7, tensor var_3265_cast_fp16_8, tensor var_3265_cast_fp16_9, tensor var_3265_cast_fp16_10, tensor var_3265_cast_fp16_11, tensor var_3265_cast_fp16_12, tensor var_3265_cast_fp16_13, tensor var_3265_cast_fp16_14, tensor var_3265_cast_fp16_15 = split(axis = var_3265_axis_0, split_sizes = tile_32, x = var_3230_cast_fp16)[name = string("op_3265_cast_fp16")]; tensor transpose_320_perm_0 = const()[name = string("transpose_320_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1604 = const()[name = string("concat_1604"), val = tensor([1, 104, 64])]; tensor transpose_320_cast_fp16 = transpose(perm = transpose_320_perm_0, x = var_3231_cast_fp16_0)[name = string("transpose_3743")]; tensor reshape_480_cast_fp16 = reshape(shape = concat_1604, x = transpose_320_cast_fp16)[name = string("reshape_480_cast_fp16")]; tensor transpose_321_perm_0 = const()[name = string("transpose_321_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1605 = const()[name = string("concat_1605"), val = tensor([1, 64, 104])]; tensor transpose_321_cast_fp16 = transpose(perm = transpose_321_perm_0, x = var_3248_cast_fp16_0)[name = string("transpose_3742")]; tensor reshape_481_cast_fp16 = reshape(shape = concat_1605, x = transpose_321_cast_fp16)[name = string("reshape_481_cast_fp16")]; bool matmul_160_transpose_x_0 = const()[name = string("matmul_160_transpose_x_0"), val = bool(false)]; bool matmul_160_transpose_y_0 = const()[name = string("matmul_160_transpose_y_0"), val = bool(false)]; tensor matmul_160_cast_fp16 = matmul(transpose_x = matmul_160_transpose_x_0, transpose_y = matmul_160_transpose_y_0, x = reshape_480_cast_fp16, y = reshape_481_cast_fp16)[name = string("matmul_160_cast_fp16")]; tensor concat_1609 = const()[name = string("concat_1609"), val = tensor([1, 1, 104, 104])]; tensor reshape_482_cast_fp16 = reshape(shape = concat_1609, x = matmul_160_cast_fp16)[name = string("reshape_482_cast_fp16")]; tensor transpose_2848_perm_0 = const()[name = string("transpose_2848_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2848 = transpose(perm = transpose_2848_perm_0, x = reshape_482_cast_fp16)[name = string("transpose_3741")]; tensor w_643_cast_fp16 = add(x = transpose_2848, y = transpose_2305)[name = string("w_643_cast_fp16")]; tensor var_3287_cast_fp16 = softmax(axis = var_3175, x = w_643_cast_fp16)[name = string("op_3287_cast_fp16")]; string var_3289_equation_0 = const()[name = string("op_3289_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3289_cast_fp16 = einsum(equation = var_3289_equation_0, values = (var_3265_cast_fp16_0, var_3287_cast_fp16))[name = string("op_3289_cast_fp16")]; tensor transpose_322_perm_0 = const()[name = string("transpose_322_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1614 = const()[name = string("concat_1614"), val = tensor([1, 104, 64])]; tensor transpose_322_cast_fp16 = transpose(perm = transpose_322_perm_0, x = var_3231_cast_fp16_1)[name = string("transpose_3740")]; tensor reshape_483_cast_fp16 = reshape(shape = concat_1614, x = transpose_322_cast_fp16)[name = string("reshape_483_cast_fp16")]; tensor transpose_323_perm_0 = const()[name = string("transpose_323_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1615 = const()[name = string("concat_1615"), val = tensor([1, 64, 104])]; tensor transpose_323_cast_fp16 = transpose(perm = transpose_323_perm_0, x = var_3248_cast_fp16_1)[name = string("transpose_3739")]; tensor reshape_484_cast_fp16 = reshape(shape = concat_1615, x = transpose_323_cast_fp16)[name = string("reshape_484_cast_fp16")]; bool matmul_161_transpose_x_0 = const()[name = string("matmul_161_transpose_x_0"), val = bool(false)]; bool matmul_161_transpose_y_0 = const()[name = string("matmul_161_transpose_y_0"), val = bool(false)]; tensor matmul_161_cast_fp16 = matmul(transpose_x = matmul_161_transpose_x_0, transpose_y = matmul_161_transpose_y_0, x = reshape_483_cast_fp16, y = reshape_484_cast_fp16)[name = string("matmul_161_cast_fp16")]; tensor concat_1619 = const()[name = string("concat_1619"), val = tensor([1, 1, 104, 104])]; tensor reshape_485_cast_fp16 = reshape(shape = concat_1619, x = matmul_161_cast_fp16)[name = string("reshape_485_cast_fp16")]; tensor transpose_2849_perm_0 = const()[name = string("transpose_2849_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2849 = transpose(perm = transpose_2849_perm_0, x = reshape_485_cast_fp16)[name = string("transpose_3738")]; tensor w_647_cast_fp16 = add(x = transpose_2849, y = transpose_2305)[name = string("w_647_cast_fp16")]; tensor var_3295_cast_fp16 = softmax(axis = var_3175, x = w_647_cast_fp16)[name = string("op_3295_cast_fp16")]; string var_3297_equation_0 = const()[name = string("op_3297_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3297_cast_fp16 = einsum(equation = var_3297_equation_0, values = (var_3265_cast_fp16_1, var_3295_cast_fp16))[name = string("op_3297_cast_fp16")]; tensor transpose_324_perm_0 = const()[name = string("transpose_324_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1624 = const()[name = string("concat_1624"), val = tensor([1, 104, 64])]; tensor transpose_324_cast_fp16 = transpose(perm = transpose_324_perm_0, x = var_3231_cast_fp16_2)[name = string("transpose_3737")]; tensor reshape_486_cast_fp16 = reshape(shape = concat_1624, x = transpose_324_cast_fp16)[name = string("reshape_486_cast_fp16")]; tensor transpose_325_perm_0 = const()[name = string("transpose_325_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1625 = const()[name = string("concat_1625"), val = tensor([1, 64, 104])]; tensor transpose_325_cast_fp16 = transpose(perm = transpose_325_perm_0, x = var_3248_cast_fp16_2)[name = string("transpose_3736")]; tensor reshape_487_cast_fp16 = reshape(shape = concat_1625, x = transpose_325_cast_fp16)[name = string("reshape_487_cast_fp16")]; bool matmul_162_transpose_x_0 = const()[name = string("matmul_162_transpose_x_0"), val = bool(false)]; bool matmul_162_transpose_y_0 = const()[name = string("matmul_162_transpose_y_0"), val = bool(false)]; tensor matmul_162_cast_fp16 = matmul(transpose_x = matmul_162_transpose_x_0, transpose_y = matmul_162_transpose_y_0, x = reshape_486_cast_fp16, y = reshape_487_cast_fp16)[name = string("matmul_162_cast_fp16")]; tensor concat_1629 = const()[name = string("concat_1629"), val = tensor([1, 1, 104, 104])]; tensor reshape_488_cast_fp16 = reshape(shape = concat_1629, x = matmul_162_cast_fp16)[name = string("reshape_488_cast_fp16")]; tensor transpose_2850_perm_0 = const()[name = string("transpose_2850_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2850 = transpose(perm = transpose_2850_perm_0, x = reshape_488_cast_fp16)[name = string("transpose_3735")]; tensor w_651_cast_fp16 = add(x = transpose_2850, y = transpose_2305)[name = string("w_651_cast_fp16")]; tensor var_3303_cast_fp16 = softmax(axis = var_3175, x = w_651_cast_fp16)[name = string("op_3303_cast_fp16")]; string var_3305_equation_0 = const()[name = string("op_3305_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3305_cast_fp16 = einsum(equation = var_3305_equation_0, values = (var_3265_cast_fp16_2, var_3303_cast_fp16))[name = string("op_3305_cast_fp16")]; tensor transpose_326_perm_0 = const()[name = string("transpose_326_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1634 = const()[name = string("concat_1634"), val = tensor([1, 104, 64])]; tensor transpose_326_cast_fp16 = transpose(perm = transpose_326_perm_0, x = var_3231_cast_fp16_3)[name = string("transpose_3734")]; tensor reshape_489_cast_fp16 = reshape(shape = concat_1634, x = transpose_326_cast_fp16)[name = string("reshape_489_cast_fp16")]; tensor transpose_327_perm_0 = const()[name = string("transpose_327_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1635 = const()[name = string("concat_1635"), val = tensor([1, 64, 104])]; tensor transpose_327_cast_fp16 = transpose(perm = transpose_327_perm_0, x = var_3248_cast_fp16_3)[name = string("transpose_3733")]; tensor reshape_490_cast_fp16 = reshape(shape = concat_1635, x = transpose_327_cast_fp16)[name = string("reshape_490_cast_fp16")]; bool matmul_163_transpose_x_0 = const()[name = string("matmul_163_transpose_x_0"), val = bool(false)]; bool matmul_163_transpose_y_0 = const()[name = string("matmul_163_transpose_y_0"), val = bool(false)]; tensor matmul_163_cast_fp16 = matmul(transpose_x = matmul_163_transpose_x_0, transpose_y = matmul_163_transpose_y_0, x = reshape_489_cast_fp16, y = reshape_490_cast_fp16)[name = string("matmul_163_cast_fp16")]; tensor concat_1639 = const()[name = string("concat_1639"), val = tensor([1, 1, 104, 104])]; tensor reshape_491_cast_fp16 = reshape(shape = concat_1639, x = matmul_163_cast_fp16)[name = string("reshape_491_cast_fp16")]; tensor transpose_2851_perm_0 = const()[name = string("transpose_2851_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2851 = transpose(perm = transpose_2851_perm_0, x = reshape_491_cast_fp16)[name = string("transpose_3732")]; tensor w_655_cast_fp16 = add(x = transpose_2851, y = transpose_2305)[name = string("w_655_cast_fp16")]; tensor var_3311_cast_fp16 = softmax(axis = var_3175, x = w_655_cast_fp16)[name = string("op_3311_cast_fp16")]; string var_3313_equation_0 = const()[name = string("op_3313_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3313_cast_fp16 = einsum(equation = var_3313_equation_0, values = (var_3265_cast_fp16_3, var_3311_cast_fp16))[name = string("op_3313_cast_fp16")]; tensor transpose_328_perm_0 = const()[name = string("transpose_328_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1644 = const()[name = string("concat_1644"), val = tensor([1, 104, 64])]; tensor transpose_328_cast_fp16 = transpose(perm = transpose_328_perm_0, x = var_3231_cast_fp16_4)[name = string("transpose_3731")]; tensor reshape_492_cast_fp16 = reshape(shape = concat_1644, x = transpose_328_cast_fp16)[name = string("reshape_492_cast_fp16")]; tensor transpose_329_perm_0 = const()[name = string("transpose_329_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1645 = const()[name = string("concat_1645"), val = tensor([1, 64, 104])]; tensor transpose_329_cast_fp16 = transpose(perm = transpose_329_perm_0, x = var_3248_cast_fp16_4)[name = string("transpose_3730")]; tensor reshape_493_cast_fp16 = reshape(shape = concat_1645, x = transpose_329_cast_fp16)[name = string("reshape_493_cast_fp16")]; bool matmul_164_transpose_x_0 = const()[name = string("matmul_164_transpose_x_0"), val = bool(false)]; bool matmul_164_transpose_y_0 = const()[name = string("matmul_164_transpose_y_0"), val = bool(false)]; tensor matmul_164_cast_fp16 = matmul(transpose_x = matmul_164_transpose_x_0, transpose_y = matmul_164_transpose_y_0, x = reshape_492_cast_fp16, y = reshape_493_cast_fp16)[name = string("matmul_164_cast_fp16")]; tensor concat_1649 = const()[name = string("concat_1649"), val = tensor([1, 1, 104, 104])]; tensor reshape_494_cast_fp16 = reshape(shape = concat_1649, x = matmul_164_cast_fp16)[name = string("reshape_494_cast_fp16")]; tensor transpose_2852_perm_0 = const()[name = string("transpose_2852_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2852 = transpose(perm = transpose_2852_perm_0, x = reshape_494_cast_fp16)[name = string("transpose_3729")]; tensor w_659_cast_fp16 = add(x = transpose_2852, y = transpose_2305)[name = string("w_659_cast_fp16")]; tensor var_3319_cast_fp16 = softmax(axis = var_3175, x = w_659_cast_fp16)[name = string("op_3319_cast_fp16")]; string var_3321_equation_0 = const()[name = string("op_3321_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3321_cast_fp16 = einsum(equation = var_3321_equation_0, values = (var_3265_cast_fp16_4, var_3319_cast_fp16))[name = string("op_3321_cast_fp16")]; tensor transpose_330_perm_0 = const()[name = string("transpose_330_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1654 = const()[name = string("concat_1654"), val = tensor([1, 104, 64])]; tensor transpose_330_cast_fp16 = transpose(perm = transpose_330_perm_0, x = var_3231_cast_fp16_5)[name = string("transpose_3728")]; tensor reshape_495_cast_fp16 = reshape(shape = concat_1654, x = transpose_330_cast_fp16)[name = string("reshape_495_cast_fp16")]; tensor transpose_331_perm_0 = const()[name = string("transpose_331_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1655 = const()[name = string("concat_1655"), val = tensor([1, 64, 104])]; tensor transpose_331_cast_fp16 = transpose(perm = transpose_331_perm_0, x = var_3248_cast_fp16_5)[name = string("transpose_3727")]; tensor reshape_496_cast_fp16 = reshape(shape = concat_1655, x = transpose_331_cast_fp16)[name = string("reshape_496_cast_fp16")]; bool matmul_165_transpose_x_0 = const()[name = string("matmul_165_transpose_x_0"), val = bool(false)]; bool matmul_165_transpose_y_0 = const()[name = string("matmul_165_transpose_y_0"), val = bool(false)]; tensor matmul_165_cast_fp16 = matmul(transpose_x = matmul_165_transpose_x_0, transpose_y = matmul_165_transpose_y_0, x = reshape_495_cast_fp16, y = reshape_496_cast_fp16)[name = string("matmul_165_cast_fp16")]; tensor concat_1659 = const()[name = string("concat_1659"), val = tensor([1, 1, 104, 104])]; tensor reshape_497_cast_fp16 = reshape(shape = concat_1659, x = matmul_165_cast_fp16)[name = string("reshape_497_cast_fp16")]; tensor transpose_2853_perm_0 = const()[name = string("transpose_2853_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2853 = transpose(perm = transpose_2853_perm_0, x = reshape_497_cast_fp16)[name = string("transpose_3726")]; tensor w_663_cast_fp16 = add(x = transpose_2853, y = transpose_2305)[name = string("w_663_cast_fp16")]; tensor var_3327_cast_fp16 = softmax(axis = var_3175, x = w_663_cast_fp16)[name = string("op_3327_cast_fp16")]; string var_3329_equation_0 = const()[name = string("op_3329_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3329_cast_fp16 = einsum(equation = var_3329_equation_0, values = (var_3265_cast_fp16_5, var_3327_cast_fp16))[name = string("op_3329_cast_fp16")]; tensor transpose_332_perm_0 = const()[name = string("transpose_332_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1664 = const()[name = string("concat_1664"), val = tensor([1, 104, 64])]; tensor transpose_332_cast_fp16 = transpose(perm = transpose_332_perm_0, x = var_3231_cast_fp16_6)[name = string("transpose_3725")]; tensor reshape_498_cast_fp16 = reshape(shape = concat_1664, x = transpose_332_cast_fp16)[name = string("reshape_498_cast_fp16")]; tensor transpose_333_perm_0 = const()[name = string("transpose_333_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1665 = const()[name = string("concat_1665"), val = tensor([1, 64, 104])]; tensor transpose_333_cast_fp16 = transpose(perm = transpose_333_perm_0, x = var_3248_cast_fp16_6)[name = string("transpose_3724")]; tensor reshape_499_cast_fp16 = reshape(shape = concat_1665, x = transpose_333_cast_fp16)[name = string("reshape_499_cast_fp16")]; bool matmul_166_transpose_x_0 = const()[name = string("matmul_166_transpose_x_0"), val = bool(false)]; bool matmul_166_transpose_y_0 = const()[name = string("matmul_166_transpose_y_0"), val = bool(false)]; tensor matmul_166_cast_fp16 = matmul(transpose_x = matmul_166_transpose_x_0, transpose_y = matmul_166_transpose_y_0, x = reshape_498_cast_fp16, y = reshape_499_cast_fp16)[name = string("matmul_166_cast_fp16")]; tensor concat_1669 = const()[name = string("concat_1669"), val = tensor([1, 1, 104, 104])]; tensor reshape_500_cast_fp16 = reshape(shape = concat_1669, x = matmul_166_cast_fp16)[name = string("reshape_500_cast_fp16")]; tensor transpose_2854_perm_0 = const()[name = string("transpose_2854_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2854 = transpose(perm = transpose_2854_perm_0, x = reshape_500_cast_fp16)[name = string("transpose_3723")]; tensor w_667_cast_fp16 = add(x = transpose_2854, y = transpose_2305)[name = string("w_667_cast_fp16")]; tensor var_3335_cast_fp16 = softmax(axis = var_3175, x = w_667_cast_fp16)[name = string("op_3335_cast_fp16")]; string var_3337_equation_0 = const()[name = string("op_3337_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3337_cast_fp16 = einsum(equation = var_3337_equation_0, values = (var_3265_cast_fp16_6, var_3335_cast_fp16))[name = string("op_3337_cast_fp16")]; tensor transpose_334_perm_0 = const()[name = string("transpose_334_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1674 = const()[name = string("concat_1674"), val = tensor([1, 104, 64])]; tensor transpose_334_cast_fp16 = transpose(perm = transpose_334_perm_0, x = var_3231_cast_fp16_7)[name = string("transpose_3722")]; tensor reshape_501_cast_fp16 = reshape(shape = concat_1674, x = transpose_334_cast_fp16)[name = string("reshape_501_cast_fp16")]; tensor transpose_335_perm_0 = const()[name = string("transpose_335_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1675 = const()[name = string("concat_1675"), val = tensor([1, 64, 104])]; tensor transpose_335_cast_fp16 = transpose(perm = transpose_335_perm_0, x = var_3248_cast_fp16_7)[name = string("transpose_3721")]; tensor reshape_502_cast_fp16 = reshape(shape = concat_1675, x = transpose_335_cast_fp16)[name = string("reshape_502_cast_fp16")]; bool matmul_167_transpose_x_0 = const()[name = string("matmul_167_transpose_x_0"), val = bool(false)]; bool matmul_167_transpose_y_0 = const()[name = string("matmul_167_transpose_y_0"), val = bool(false)]; tensor matmul_167_cast_fp16 = matmul(transpose_x = matmul_167_transpose_x_0, transpose_y = matmul_167_transpose_y_0, x = reshape_501_cast_fp16, y = reshape_502_cast_fp16)[name = string("matmul_167_cast_fp16")]; tensor concat_1679 = const()[name = string("concat_1679"), val = tensor([1, 1, 104, 104])]; tensor reshape_503_cast_fp16 = reshape(shape = concat_1679, x = matmul_167_cast_fp16)[name = string("reshape_503_cast_fp16")]; tensor transpose_2855_perm_0 = const()[name = string("transpose_2855_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2855 = transpose(perm = transpose_2855_perm_0, x = reshape_503_cast_fp16)[name = string("transpose_3720")]; tensor w_671_cast_fp16 = add(x = transpose_2855, y = transpose_2305)[name = string("w_671_cast_fp16")]; tensor var_3343_cast_fp16 = softmax(axis = var_3175, x = w_671_cast_fp16)[name = string("op_3343_cast_fp16")]; string var_3345_equation_0 = const()[name = string("op_3345_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3345_cast_fp16 = einsum(equation = var_3345_equation_0, values = (var_3265_cast_fp16_7, var_3343_cast_fp16))[name = string("op_3345_cast_fp16")]; tensor transpose_336_perm_0 = const()[name = string("transpose_336_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1684 = const()[name = string("concat_1684"), val = tensor([1, 104, 64])]; tensor transpose_336_cast_fp16 = transpose(perm = transpose_336_perm_0, x = var_3231_cast_fp16_8)[name = string("transpose_3719")]; tensor reshape_504_cast_fp16 = reshape(shape = concat_1684, x = transpose_336_cast_fp16)[name = string("reshape_504_cast_fp16")]; tensor transpose_337_perm_0 = const()[name = string("transpose_337_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1685 = const()[name = string("concat_1685"), val = tensor([1, 64, 104])]; tensor transpose_337_cast_fp16 = transpose(perm = transpose_337_perm_0, x = var_3248_cast_fp16_8)[name = string("transpose_3718")]; tensor reshape_505_cast_fp16 = reshape(shape = concat_1685, x = transpose_337_cast_fp16)[name = string("reshape_505_cast_fp16")]; bool matmul_168_transpose_x_0 = const()[name = string("matmul_168_transpose_x_0"), val = bool(false)]; bool matmul_168_transpose_y_0 = const()[name = string("matmul_168_transpose_y_0"), val = bool(false)]; tensor matmul_168_cast_fp16 = matmul(transpose_x = matmul_168_transpose_x_0, transpose_y = matmul_168_transpose_y_0, x = reshape_504_cast_fp16, y = reshape_505_cast_fp16)[name = string("matmul_168_cast_fp16")]; tensor concat_1689 = const()[name = string("concat_1689"), val = tensor([1, 1, 104, 104])]; tensor reshape_506_cast_fp16 = reshape(shape = concat_1689, x = matmul_168_cast_fp16)[name = string("reshape_506_cast_fp16")]; tensor transpose_2856_perm_0 = const()[name = string("transpose_2856_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2856 = transpose(perm = transpose_2856_perm_0, x = reshape_506_cast_fp16)[name = string("transpose_3717")]; tensor w_675_cast_fp16 = add(x = transpose_2856, y = transpose_2305)[name = string("w_675_cast_fp16")]; tensor var_3351_cast_fp16 = softmax(axis = var_3175, x = w_675_cast_fp16)[name = string("op_3351_cast_fp16")]; string var_3353_equation_0 = const()[name = string("op_3353_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3353_cast_fp16 = einsum(equation = var_3353_equation_0, values = (var_3265_cast_fp16_8, var_3351_cast_fp16))[name = string("op_3353_cast_fp16")]; tensor transpose_338_perm_0 = const()[name = string("transpose_338_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1694 = const()[name = string("concat_1694"), val = tensor([1, 104, 64])]; tensor transpose_338_cast_fp16 = transpose(perm = transpose_338_perm_0, x = var_3231_cast_fp16_9)[name = string("transpose_3716")]; tensor reshape_507_cast_fp16 = reshape(shape = concat_1694, x = transpose_338_cast_fp16)[name = string("reshape_507_cast_fp16")]; tensor transpose_339_perm_0 = const()[name = string("transpose_339_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1695 = const()[name = string("concat_1695"), val = tensor([1, 64, 104])]; tensor transpose_339_cast_fp16 = transpose(perm = transpose_339_perm_0, x = var_3248_cast_fp16_9)[name = string("transpose_3715")]; tensor reshape_508_cast_fp16 = reshape(shape = concat_1695, x = transpose_339_cast_fp16)[name = string("reshape_508_cast_fp16")]; bool matmul_169_transpose_x_0 = const()[name = string("matmul_169_transpose_x_0"), val = bool(false)]; bool matmul_169_transpose_y_0 = const()[name = string("matmul_169_transpose_y_0"), val = bool(false)]; tensor matmul_169_cast_fp16 = matmul(transpose_x = matmul_169_transpose_x_0, transpose_y = matmul_169_transpose_y_0, x = reshape_507_cast_fp16, y = reshape_508_cast_fp16)[name = string("matmul_169_cast_fp16")]; tensor concat_1699 = const()[name = string("concat_1699"), val = tensor([1, 1, 104, 104])]; tensor reshape_509_cast_fp16 = reshape(shape = concat_1699, x = matmul_169_cast_fp16)[name = string("reshape_509_cast_fp16")]; tensor transpose_2857_perm_0 = const()[name = string("transpose_2857_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2857 = transpose(perm = transpose_2857_perm_0, x = reshape_509_cast_fp16)[name = string("transpose_3714")]; tensor w_679_cast_fp16 = add(x = transpose_2857, y = transpose_2305)[name = string("w_679_cast_fp16")]; tensor var_3359_cast_fp16 = softmax(axis = var_3175, x = w_679_cast_fp16)[name = string("op_3359_cast_fp16")]; string var_3361_equation_0 = const()[name = string("op_3361_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3361_cast_fp16 = einsum(equation = var_3361_equation_0, values = (var_3265_cast_fp16_9, var_3359_cast_fp16))[name = string("op_3361_cast_fp16")]; tensor transpose_340_perm_0 = const()[name = string("transpose_340_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1704 = const()[name = string("concat_1704"), val = tensor([1, 104, 64])]; tensor transpose_340_cast_fp16 = transpose(perm = transpose_340_perm_0, x = var_3231_cast_fp16_10)[name = string("transpose_3713")]; tensor reshape_510_cast_fp16 = reshape(shape = concat_1704, x = transpose_340_cast_fp16)[name = string("reshape_510_cast_fp16")]; tensor transpose_341_perm_0 = const()[name = string("transpose_341_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1705 = const()[name = string("concat_1705"), val = tensor([1, 64, 104])]; tensor transpose_341_cast_fp16 = transpose(perm = transpose_341_perm_0, x = var_3248_cast_fp16_10)[name = string("transpose_3712")]; tensor reshape_511_cast_fp16 = reshape(shape = concat_1705, x = transpose_341_cast_fp16)[name = string("reshape_511_cast_fp16")]; bool matmul_170_transpose_x_0 = const()[name = string("matmul_170_transpose_x_0"), val = bool(false)]; bool matmul_170_transpose_y_0 = const()[name = string("matmul_170_transpose_y_0"), val = bool(false)]; tensor matmul_170_cast_fp16 = matmul(transpose_x = matmul_170_transpose_x_0, transpose_y = matmul_170_transpose_y_0, x = reshape_510_cast_fp16, y = reshape_511_cast_fp16)[name = string("matmul_170_cast_fp16")]; tensor concat_1709 = const()[name = string("concat_1709"), val = tensor([1, 1, 104, 104])]; tensor reshape_512_cast_fp16 = reshape(shape = concat_1709, x = matmul_170_cast_fp16)[name = string("reshape_512_cast_fp16")]; tensor transpose_2858_perm_0 = const()[name = string("transpose_2858_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2858 = transpose(perm = transpose_2858_perm_0, x = reshape_512_cast_fp16)[name = string("transpose_3711")]; tensor w_683_cast_fp16 = add(x = transpose_2858, y = transpose_2305)[name = string("w_683_cast_fp16")]; tensor var_3367_cast_fp16 = softmax(axis = var_3175, x = w_683_cast_fp16)[name = string("op_3367_cast_fp16")]; string var_3369_equation_0 = const()[name = string("op_3369_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3369_cast_fp16 = einsum(equation = var_3369_equation_0, values = (var_3265_cast_fp16_10, var_3367_cast_fp16))[name = string("op_3369_cast_fp16")]; tensor transpose_342_perm_0 = const()[name = string("transpose_342_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1714 = const()[name = string("concat_1714"), val = tensor([1, 104, 64])]; tensor transpose_342_cast_fp16 = transpose(perm = transpose_342_perm_0, x = var_3231_cast_fp16_11)[name = string("transpose_3710")]; tensor reshape_513_cast_fp16 = reshape(shape = concat_1714, x = transpose_342_cast_fp16)[name = string("reshape_513_cast_fp16")]; tensor transpose_343_perm_0 = const()[name = string("transpose_343_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1715 = const()[name = string("concat_1715"), val = tensor([1, 64, 104])]; tensor transpose_343_cast_fp16 = transpose(perm = transpose_343_perm_0, x = var_3248_cast_fp16_11)[name = string("transpose_3709")]; tensor reshape_514_cast_fp16 = reshape(shape = concat_1715, x = transpose_343_cast_fp16)[name = string("reshape_514_cast_fp16")]; bool matmul_171_transpose_x_0 = const()[name = string("matmul_171_transpose_x_0"), val = bool(false)]; bool matmul_171_transpose_y_0 = const()[name = string("matmul_171_transpose_y_0"), val = bool(false)]; tensor matmul_171_cast_fp16 = matmul(transpose_x = matmul_171_transpose_x_0, transpose_y = matmul_171_transpose_y_0, x = reshape_513_cast_fp16, y = reshape_514_cast_fp16)[name = string("matmul_171_cast_fp16")]; tensor concat_1719 = const()[name = string("concat_1719"), val = tensor([1, 1, 104, 104])]; tensor reshape_515_cast_fp16 = reshape(shape = concat_1719, x = matmul_171_cast_fp16)[name = string("reshape_515_cast_fp16")]; tensor transpose_2859_perm_0 = const()[name = string("transpose_2859_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2859 = transpose(perm = transpose_2859_perm_0, x = reshape_515_cast_fp16)[name = string("transpose_3708")]; tensor w_687_cast_fp16 = add(x = transpose_2859, y = transpose_2305)[name = string("w_687_cast_fp16")]; tensor var_3375_cast_fp16 = softmax(axis = var_3175, x = w_687_cast_fp16)[name = string("op_3375_cast_fp16")]; string var_3377_equation_0 = const()[name = string("op_3377_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3377_cast_fp16 = einsum(equation = var_3377_equation_0, values = (var_3265_cast_fp16_11, var_3375_cast_fp16))[name = string("op_3377_cast_fp16")]; tensor transpose_344_perm_0 = const()[name = string("transpose_344_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1724 = const()[name = string("concat_1724"), val = tensor([1, 104, 64])]; tensor transpose_344_cast_fp16 = transpose(perm = transpose_344_perm_0, x = var_3231_cast_fp16_12)[name = string("transpose_3707")]; tensor reshape_516_cast_fp16 = reshape(shape = concat_1724, x = transpose_344_cast_fp16)[name = string("reshape_516_cast_fp16")]; tensor transpose_345_perm_0 = const()[name = string("transpose_345_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1725 = const()[name = string("concat_1725"), val = tensor([1, 64, 104])]; tensor transpose_345_cast_fp16 = transpose(perm = transpose_345_perm_0, x = var_3248_cast_fp16_12)[name = string("transpose_3706")]; tensor reshape_517_cast_fp16 = reshape(shape = concat_1725, x = transpose_345_cast_fp16)[name = string("reshape_517_cast_fp16")]; bool matmul_172_transpose_x_0 = const()[name = string("matmul_172_transpose_x_0"), val = bool(false)]; bool matmul_172_transpose_y_0 = const()[name = string("matmul_172_transpose_y_0"), val = bool(false)]; tensor matmul_172_cast_fp16 = matmul(transpose_x = matmul_172_transpose_x_0, transpose_y = matmul_172_transpose_y_0, x = reshape_516_cast_fp16, y = reshape_517_cast_fp16)[name = string("matmul_172_cast_fp16")]; tensor concat_1729 = const()[name = string("concat_1729"), val = tensor([1, 1, 104, 104])]; tensor reshape_518_cast_fp16 = reshape(shape = concat_1729, x = matmul_172_cast_fp16)[name = string("reshape_518_cast_fp16")]; tensor transpose_2860_perm_0 = const()[name = string("transpose_2860_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2860 = transpose(perm = transpose_2860_perm_0, x = reshape_518_cast_fp16)[name = string("transpose_3705")]; tensor w_691_cast_fp16 = add(x = transpose_2860, y = transpose_2305)[name = string("w_691_cast_fp16")]; tensor var_3383_cast_fp16 = softmax(axis = var_3175, x = w_691_cast_fp16)[name = string("op_3383_cast_fp16")]; string var_3385_equation_0 = const()[name = string("op_3385_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3385_cast_fp16 = einsum(equation = var_3385_equation_0, values = (var_3265_cast_fp16_12, var_3383_cast_fp16))[name = string("op_3385_cast_fp16")]; tensor transpose_346_perm_0 = const()[name = string("transpose_346_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1734 = const()[name = string("concat_1734"), val = tensor([1, 104, 64])]; tensor transpose_346_cast_fp16 = transpose(perm = transpose_346_perm_0, x = var_3231_cast_fp16_13)[name = string("transpose_3704")]; tensor reshape_519_cast_fp16 = reshape(shape = concat_1734, x = transpose_346_cast_fp16)[name = string("reshape_519_cast_fp16")]; tensor transpose_347_perm_0 = const()[name = string("transpose_347_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1735 = const()[name = string("concat_1735"), val = tensor([1, 64, 104])]; tensor transpose_347_cast_fp16 = transpose(perm = transpose_347_perm_0, x = var_3248_cast_fp16_13)[name = string("transpose_3703")]; tensor reshape_520_cast_fp16 = reshape(shape = concat_1735, x = transpose_347_cast_fp16)[name = string("reshape_520_cast_fp16")]; bool matmul_173_transpose_x_0 = const()[name = string("matmul_173_transpose_x_0"), val = bool(false)]; bool matmul_173_transpose_y_0 = const()[name = string("matmul_173_transpose_y_0"), val = bool(false)]; tensor matmul_173_cast_fp16 = matmul(transpose_x = matmul_173_transpose_x_0, transpose_y = matmul_173_transpose_y_0, x = reshape_519_cast_fp16, y = reshape_520_cast_fp16)[name = string("matmul_173_cast_fp16")]; tensor concat_1739 = const()[name = string("concat_1739"), val = tensor([1, 1, 104, 104])]; tensor reshape_521_cast_fp16 = reshape(shape = concat_1739, x = matmul_173_cast_fp16)[name = string("reshape_521_cast_fp16")]; tensor transpose_2861_perm_0 = const()[name = string("transpose_2861_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2861 = transpose(perm = transpose_2861_perm_0, x = reshape_521_cast_fp16)[name = string("transpose_3702")]; tensor w_695_cast_fp16 = add(x = transpose_2861, y = transpose_2305)[name = string("w_695_cast_fp16")]; tensor var_3391_cast_fp16 = softmax(axis = var_3175, x = w_695_cast_fp16)[name = string("op_3391_cast_fp16")]; string var_3393_equation_0 = const()[name = string("op_3393_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3393_cast_fp16 = einsum(equation = var_3393_equation_0, values = (var_3265_cast_fp16_13, var_3391_cast_fp16))[name = string("op_3393_cast_fp16")]; tensor transpose_348_perm_0 = const()[name = string("transpose_348_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1744 = const()[name = string("concat_1744"), val = tensor([1, 104, 64])]; tensor transpose_348_cast_fp16 = transpose(perm = transpose_348_perm_0, x = var_3231_cast_fp16_14)[name = string("transpose_3701")]; tensor reshape_522_cast_fp16 = reshape(shape = concat_1744, x = transpose_348_cast_fp16)[name = string("reshape_522_cast_fp16")]; tensor transpose_349_perm_0 = const()[name = string("transpose_349_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1745 = const()[name = string("concat_1745"), val = tensor([1, 64, 104])]; tensor transpose_349_cast_fp16 = transpose(perm = transpose_349_perm_0, x = var_3248_cast_fp16_14)[name = string("transpose_3700")]; tensor reshape_523_cast_fp16 = reshape(shape = concat_1745, x = transpose_349_cast_fp16)[name = string("reshape_523_cast_fp16")]; bool matmul_174_transpose_x_0 = const()[name = string("matmul_174_transpose_x_0"), val = bool(false)]; bool matmul_174_transpose_y_0 = const()[name = string("matmul_174_transpose_y_0"), val = bool(false)]; tensor matmul_174_cast_fp16 = matmul(transpose_x = matmul_174_transpose_x_0, transpose_y = matmul_174_transpose_y_0, x = reshape_522_cast_fp16, y = reshape_523_cast_fp16)[name = string("matmul_174_cast_fp16")]; tensor concat_1749 = const()[name = string("concat_1749"), val = tensor([1, 1, 104, 104])]; tensor reshape_524_cast_fp16 = reshape(shape = concat_1749, x = matmul_174_cast_fp16)[name = string("reshape_524_cast_fp16")]; tensor transpose_2862_perm_0 = const()[name = string("transpose_2862_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2862 = transpose(perm = transpose_2862_perm_0, x = reshape_524_cast_fp16)[name = string("transpose_3699")]; tensor w_699_cast_fp16 = add(x = transpose_2862, y = transpose_2305)[name = string("w_699_cast_fp16")]; tensor var_3399_cast_fp16 = softmax(axis = var_3175, x = w_699_cast_fp16)[name = string("op_3399_cast_fp16")]; string var_3401_equation_0 = const()[name = string("op_3401_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3401_cast_fp16 = einsum(equation = var_3401_equation_0, values = (var_3265_cast_fp16_14, var_3399_cast_fp16))[name = string("op_3401_cast_fp16")]; tensor transpose_350_perm_0 = const()[name = string("transpose_350_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1754 = const()[name = string("concat_1754"), val = tensor([1, 104, 64])]; tensor transpose_350_cast_fp16 = transpose(perm = transpose_350_perm_0, x = var_3231_cast_fp16_15)[name = string("transpose_3698")]; tensor reshape_525_cast_fp16 = reshape(shape = concat_1754, x = transpose_350_cast_fp16)[name = string("reshape_525_cast_fp16")]; tensor transpose_351_perm_0 = const()[name = string("transpose_351_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1755 = const()[name = string("concat_1755"), val = tensor([1, 64, 104])]; tensor transpose_351_cast_fp16 = transpose(perm = transpose_351_perm_0, x = var_3248_cast_fp16_15)[name = string("transpose_3697")]; tensor reshape_526_cast_fp16 = reshape(shape = concat_1755, x = transpose_351_cast_fp16)[name = string("reshape_526_cast_fp16")]; bool matmul_175_transpose_x_0 = const()[name = string("matmul_175_transpose_x_0"), val = bool(false)]; bool matmul_175_transpose_y_0 = const()[name = string("matmul_175_transpose_y_0"), val = bool(false)]; tensor matmul_175_cast_fp16 = matmul(transpose_x = matmul_175_transpose_x_0, transpose_y = matmul_175_transpose_y_0, x = reshape_525_cast_fp16, y = reshape_526_cast_fp16)[name = string("matmul_175_cast_fp16")]; tensor concat_1759 = const()[name = string("concat_1759"), val = tensor([1, 1, 104, 104])]; tensor reshape_527_cast_fp16 = reshape(shape = concat_1759, x = matmul_175_cast_fp16)[name = string("reshape_527_cast_fp16")]; tensor transpose_2863_perm_0 = const()[name = string("transpose_2863_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2863 = transpose(perm = transpose_2863_perm_0, x = reshape_527_cast_fp16)[name = string("transpose_3696")]; tensor w_703_cast_fp16 = add(x = transpose_2863, y = transpose_2305)[name = string("w_703_cast_fp16")]; tensor var_3407_cast_fp16 = softmax(axis = var_3175, x = w_703_cast_fp16)[name = string("op_3407_cast_fp16")]; string var_3409_equation_0 = const()[name = string("op_3409_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3409_cast_fp16 = einsum(equation = var_3409_equation_0, values = (var_3265_cast_fp16_15, var_3407_cast_fp16))[name = string("op_3409_cast_fp16")]; bool input_91_interleave_0 = const()[name = string("input_91_interleave_0"), val = bool(false)]; tensor input_91_cast_fp16 = concat(axis = var_3175, interleave = input_91_interleave_0, values = (var_3289_cast_fp16, var_3297_cast_fp16, var_3305_cast_fp16, var_3313_cast_fp16, var_3321_cast_fp16, var_3329_cast_fp16, var_3337_cast_fp16, var_3345_cast_fp16, var_3353_cast_fp16, var_3361_cast_fp16, var_3369_cast_fp16, var_3377_cast_fp16, var_3385_cast_fp16, var_3393_cast_fp16, var_3401_cast_fp16, var_3409_cast_fp16))[name = string("input_91_cast_fp16")]; string var_3418_pad_type_0 = const()[name = string("op_3418_pad_type_0"), val = string("valid")]; tensor var_3418_strides_0 = const()[name = string("op_3418_strides_0"), val = tensor([1, 1])]; tensor var_3418_pad_0 = const()[name = string("op_3418_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3418_dilations_0 = const()[name = string("op_3418_dilations_0"), val = tensor([1, 1])]; int32 var_3418_groups_0 = const()[name = string("op_3418_groups_0"), val = int32(1)]; tensor layers_10_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_10_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(282307136)))]; tensor layers_10_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_10_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(284404352)))]; tensor var_3418_cast_fp16 = conv(bias = layers_10_self_attn_out_proj_bias_to_fp16, dilations = var_3418_dilations_0, groups = var_3418_groups_0, pad = var_3418_pad_0, pad_type = var_3418_pad_type_0, strides = var_3418_strides_0, weight = layers_10_self_attn_out_proj_weight_to_fp16, x = input_91_cast_fp16)[name = string("op_3418_cast_fp16")]; tensor x_117_cast_fp16 = add(x = x_113_cast_fp16, y = var_3418_cast_fp16)[name = string("x_117_cast_fp16")]; tensor mu_43_axes_0 = const()[name = string("mu_43_axes_0"), val = tensor([1])]; bool mu_43_keep_dims_0 = const()[name = string("mu_43_keep_dims_0"), val = bool(true)]; tensor mu_43_cast_fp16 = reduce_mean(axes = mu_43_axes_0, keep_dims = mu_43_keep_dims_0, x = x_117_cast_fp16)[name = string("mu_43_cast_fp16")]; tensor var_3424_cast_fp16 = sub(x = x_117_cast_fp16, y = mu_43_cast_fp16)[name = string("op_3424_cast_fp16")]; fp16 var_3178_promoted_1_to_fp16 = const()[name = string("op_3178_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_3425_cast_fp16 = pow(x = var_3424_cast_fp16, y = var_3178_promoted_1_to_fp16)[name = string("op_3425_cast_fp16")]; tensor var_43_axes_0 = const()[name = string("var_43_axes_0"), val = tensor([1])]; bool var_43_keep_dims_0 = const()[name = string("var_43_keep_dims_0"), val = bool(true)]; tensor var_43_cast_fp16 = reduce_mean(axes = var_43_axes_0, keep_dims = var_43_keep_dims_0, x = var_3425_cast_fp16)[name = string("var_43_cast_fp16")]; fp16 var_3429_to_fp16 = const()[name = string("op_3429_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_3430_cast_fp16 = add(x = var_43_cast_fp16, y = var_3429_to_fp16)[name = string("op_3430_cast_fp16")]; fp32 var_3431_epsilon_0 = const()[name = string("op_3431_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_3431_cast_fp16 = rsqrt(epsilon = var_3431_epsilon_0, x = var_3430_cast_fp16)[name = string("op_3431_cast_fp16")]; tensor x_119_cast_fp16 = mul(x = var_3424_cast_fp16, y = var_3431_cast_fp16)[name = string("x_119_cast_fp16")]; tensor input_93_gamma_0_to_fp16 = const()[name = string("input_93_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(284406464)))]; tensor input_93_beta_0_to_fp16 = const()[name = string("input_93_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(284408576)))]; fp16 input_93_epsilon_0_to_fp16 = const()[name = string("input_93_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_93_cast_fp16 = batch_norm(beta = input_93_beta_0_to_fp16, epsilon = input_93_epsilon_0_to_fp16, gamma = input_93_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_119_cast_fp16)[name = string("input_93_cast_fp16")]; string x_121_pad_type_0 = const()[name = string("x_121_pad_type_0"), val = string("valid")]; tensor x_121_strides_0 = const()[name = string("x_121_strides_0"), val = tensor([1, 1])]; tensor x_121_pad_0 = const()[name = string("x_121_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_121_dilations_0 = const()[name = string("x_121_dilations_0"), val = tensor([1, 1])]; int32 x_121_groups_0 = const()[name = string("x_121_groups_0"), val = int32(1)]; tensor layers_10_fc1_weight_to_fp16 = const()[name = string("layers_10_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(284410688)))]; tensor layers_10_fc1_bias_to_fp16 = const()[name = string("layers_10_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(292799360)))]; tensor x_121_cast_fp16 = conv(bias = layers_10_fc1_bias_to_fp16, dilations = x_121_dilations_0, groups = x_121_groups_0, pad = x_121_pad_0, pad_type = x_121_pad_type_0, strides = x_121_strides_0, weight = layers_10_fc1_weight_to_fp16, x = input_93_cast_fp16)[name = string("x_121_cast_fp16")]; fp16 var_3446_to_fp16 = const()[name = string("op_3446_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_3447_cast_fp16 = mul(x = x_121_cast_fp16, y = var_3446_to_fp16)[name = string("op_3447_cast_fp16")]; tensor var_3448_cast_fp16 = mul(x = var_3447_cast_fp16, y = x_121_cast_fp16)[name = string("op_3448_cast_fp16")]; tensor var_3449_cast_fp16 = mul(x = var_3448_cast_fp16, y = x_121_cast_fp16)[name = string("op_3449_cast_fp16")]; tensor var_3450_cast_fp16 = add(x = x_121_cast_fp16, y = var_3449_cast_fp16)[name = string("op_3450_cast_fp16")]; fp16 var_3451_to_fp16 = const()[name = string("op_3451_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_27_cast_fp16 = mul(x = var_3450_cast_fp16, y = var_3451_to_fp16)[name = string("u_27_cast_fp16")]; fp16 var_3453_to_fp16 = const()[name = string("op_3453_to_fp16"), val = fp16(0x1p-1)]; tensor var_3454_cast_fp16 = mul(x = x_121_cast_fp16, y = var_3453_to_fp16)[name = string("op_3454_cast_fp16")]; tensor var_3455_cast_fp16 = tanh(x = u_27_cast_fp16)[name = string("op_3455_cast_fp16")]; fp16 var_3456_to_fp16 = const()[name = string("op_3456_to_fp16"), val = fp16(0x1p+0)]; tensor var_3457_cast_fp16 = add(x = var_3455_cast_fp16, y = var_3456_to_fp16)[name = string("op_3457_cast_fp16")]; tensor input_95_cast_fp16 = mul(x = var_3454_cast_fp16, y = var_3457_cast_fp16)[name = string("input_95_cast_fp16")]; string h_21_pad_type_0 = const()[name = string("h_21_pad_type_0"), val = string("valid")]; tensor h_21_strides_0 = const()[name = string("h_21_strides_0"), val = tensor([1, 1])]; tensor h_21_pad_0 = const()[name = string("h_21_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_21_dilations_0 = const()[name = string("h_21_dilations_0"), val = tensor([1, 1])]; int32 h_21_groups_0 = const()[name = string("h_21_groups_0"), val = int32(1)]; tensor layers_10_fc2_weight_to_fp16 = const()[name = string("layers_10_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(292807616)))]; tensor layers_10_fc2_bias_to_fp16 = const()[name = string("layers_10_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(301196288)))]; tensor h_21_cast_fp16 = conv(bias = layers_10_fc2_bias_to_fp16, dilations = h_21_dilations_0, groups = h_21_groups_0, pad = h_21_pad_0, pad_type = h_21_pad_type_0, strides = h_21_strides_0, weight = layers_10_fc2_weight_to_fp16, x = input_95_cast_fp16)[name = string("h_21_cast_fp16")]; tensor x_123_cast_fp16 = add(x = x_117_cast_fp16, y = h_21_cast_fp16)[name = string("x_123_cast_fp16")]; int32 var_3473 = const()[name = string("op_3473"), val = int32(1)]; tensor mu_45_axes_0 = const()[name = string("mu_45_axes_0"), val = tensor([1])]; bool mu_45_keep_dims_0 = const()[name = string("mu_45_keep_dims_0"), val = bool(true)]; tensor mu_45_cast_fp16 = reduce_mean(axes = mu_45_axes_0, keep_dims = mu_45_keep_dims_0, x = x_123_cast_fp16)[name = string("mu_45_cast_fp16")]; tensor var_3487_cast_fp16 = sub(x = x_123_cast_fp16, y = mu_45_cast_fp16)[name = string("op_3487_cast_fp16")]; fp16 var_3476_promoted_to_fp16 = const()[name = string("op_3476_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_3488_cast_fp16 = pow(x = var_3487_cast_fp16, y = var_3476_promoted_to_fp16)[name = string("op_3488_cast_fp16")]; tensor var_45_axes_0 = const()[name = string("var_45_axes_0"), val = tensor([1])]; bool var_45_keep_dims_0 = const()[name = string("var_45_keep_dims_0"), val = bool(true)]; tensor var_45_cast_fp16 = reduce_mean(axes = var_45_axes_0, keep_dims = var_45_keep_dims_0, x = var_3488_cast_fp16)[name = string("var_45_cast_fp16")]; fp16 var_3492_to_fp16 = const()[name = string("op_3492_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_3493_cast_fp16 = add(x = var_45_cast_fp16, y = var_3492_to_fp16)[name = string("op_3493_cast_fp16")]; fp32 var_3494_epsilon_0 = const()[name = string("op_3494_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_3494_cast_fp16 = rsqrt(epsilon = var_3494_epsilon_0, x = var_3493_cast_fp16)[name = string("op_3494_cast_fp16")]; tensor x_125_cast_fp16 = mul(x = var_3487_cast_fp16, y = var_3494_cast_fp16)[name = string("x_125_cast_fp16")]; tensor input_97_gamma_0_to_fp16 = const()[name = string("input_97_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(301198400)))]; tensor input_97_beta_0_to_fp16 = const()[name = string("input_97_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(301200512)))]; fp16 input_97_epsilon_0_to_fp16 = const()[name = string("input_97_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_97_cast_fp16 = batch_norm(beta = input_97_beta_0_to_fp16, epsilon = input_97_epsilon_0_to_fp16, gamma = input_97_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_125_cast_fp16)[name = string("input_97_cast_fp16")]; string var_3512_pad_type_0 = const()[name = string("op_3512_pad_type_0"), val = string("valid")]; tensor var_3512_strides_0 = const()[name = string("op_3512_strides_0"), val = tensor([1, 1])]; tensor var_3512_pad_0 = const()[name = string("op_3512_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3512_dilations_0 = const()[name = string("op_3512_dilations_0"), val = tensor([1, 1])]; int32 var_3512_groups_0 = const()[name = string("op_3512_groups_0"), val = int32(1)]; tensor var_3514_weight_0_to_fp16 = const()[name = string("op_3514_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(301202624)))]; tensor var_3514_bias_0_to_fp16 = const()[name = string("op_3514_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(303299840)))]; tensor var_3514_cast_fp16 = conv(bias = var_3514_bias_0_to_fp16, dilations = var_3512_dilations_0, groups = var_3512_groups_0, pad = var_3512_pad_0, pad_type = var_3512_pad_type_0, strides = var_3512_strides_0, weight = var_3514_weight_0_to_fp16, x = input_97_cast_fp16)[name = string("op_3514_cast_fp16")]; string var_3521_pad_type_0 = const()[name = string("op_3521_pad_type_0"), val = string("valid")]; tensor var_3521_strides_0 = const()[name = string("op_3521_strides_0"), val = tensor([1, 1])]; tensor var_3521_pad_0 = const()[name = string("op_3521_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3521_dilations_0 = const()[name = string("op_3521_dilations_0"), val = tensor([1, 1])]; int32 var_3521_groups_0 = const()[name = string("op_3521_groups_0"), val = int32(1)]; tensor layers_11_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_11_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(303301952)))]; tensor layers_11_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_11_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(305399168)))]; tensor var_3521_cast_fp16 = conv(bias = layers_11_self_attn_k_proj_bias_to_fp16, dilations = var_3521_dilations_0, groups = var_3521_groups_0, pad = var_3521_pad_0, pad_type = var_3521_pad_type_0, strides = var_3521_strides_0, weight = layers_11_self_attn_k_proj_weight_to_fp16, x = input_97_cast_fp16)[name = string("op_3521_cast_fp16")]; string var_3528_pad_type_0 = const()[name = string("op_3528_pad_type_0"), val = string("valid")]; tensor var_3528_strides_0 = const()[name = string("op_3528_strides_0"), val = tensor([1, 1])]; tensor var_3528_pad_0 = const()[name = string("op_3528_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3528_dilations_0 = const()[name = string("op_3528_dilations_0"), val = tensor([1, 1])]; int32 var_3528_groups_0 = const()[name = string("op_3528_groups_0"), val = int32(1)]; tensor layers_11_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_11_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(305401280)))]; tensor layers_11_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_11_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(307498496)))]; tensor var_3528_cast_fp16 = conv(bias = layers_11_self_attn_v_proj_bias_to_fp16, dilations = var_3528_dilations_0, groups = var_3528_groups_0, pad = var_3528_pad_0, pad_type = var_3528_pad_type_0, strides = var_3528_strides_0, weight = layers_11_self_attn_v_proj_weight_to_fp16, x = input_97_cast_fp16)[name = string("op_3528_cast_fp16")]; tensor tile_33 = const()[name = string("tile_33"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(307500608)))]; int32 var_3529_axis_0 = const()[name = string("op_3529_axis_0"), val = int32(1)]; tensor var_3529_cast_fp16_0, tensor var_3529_cast_fp16_1, tensor var_3529_cast_fp16_2, tensor var_3529_cast_fp16_3, tensor var_3529_cast_fp16_4, tensor var_3529_cast_fp16_5, tensor var_3529_cast_fp16_6, tensor var_3529_cast_fp16_7, tensor var_3529_cast_fp16_8, tensor var_3529_cast_fp16_9, tensor var_3529_cast_fp16_10, tensor var_3529_cast_fp16_11, tensor var_3529_cast_fp16_12, tensor var_3529_cast_fp16_13, tensor var_3529_cast_fp16_14, tensor var_3529_cast_fp16_15 = split(axis = var_3529_axis_0, split_sizes = tile_33, x = var_3514_cast_fp16)[name = string("op_3529_cast_fp16")]; tensor tile_34 = const()[name = string("tile_34"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(307500736)))]; int32 var_3546_axis_0 = const()[name = string("op_3546_axis_0"), val = int32(1)]; tensor var_3546_cast_fp16_0, tensor var_3546_cast_fp16_1, tensor var_3546_cast_fp16_2, tensor var_3546_cast_fp16_3, tensor var_3546_cast_fp16_4, tensor var_3546_cast_fp16_5, tensor var_3546_cast_fp16_6, tensor var_3546_cast_fp16_7, tensor var_3546_cast_fp16_8, tensor var_3546_cast_fp16_9, tensor var_3546_cast_fp16_10, tensor var_3546_cast_fp16_11, tensor var_3546_cast_fp16_12, tensor var_3546_cast_fp16_13, tensor var_3546_cast_fp16_14, tensor var_3546_cast_fp16_15 = split(axis = var_3546_axis_0, split_sizes = tile_34, x = var_3521_cast_fp16)[name = string("op_3546_cast_fp16")]; tensor tile_35 = const()[name = string("tile_35"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(307500864)))]; int32 var_3563_axis_0 = const()[name = string("op_3563_axis_0"), val = int32(1)]; tensor var_3563_cast_fp16_0, tensor var_3563_cast_fp16_1, tensor var_3563_cast_fp16_2, tensor var_3563_cast_fp16_3, tensor var_3563_cast_fp16_4, tensor var_3563_cast_fp16_5, tensor var_3563_cast_fp16_6, tensor var_3563_cast_fp16_7, tensor var_3563_cast_fp16_8, tensor var_3563_cast_fp16_9, tensor var_3563_cast_fp16_10, tensor var_3563_cast_fp16_11, tensor var_3563_cast_fp16_12, tensor var_3563_cast_fp16_13, tensor var_3563_cast_fp16_14, tensor var_3563_cast_fp16_15 = split(axis = var_3563_axis_0, split_sizes = tile_35, x = var_3528_cast_fp16)[name = string("op_3563_cast_fp16")]; tensor transpose_352_perm_0 = const()[name = string("transpose_352_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1764 = const()[name = string("concat_1764"), val = tensor([1, 104, 64])]; tensor transpose_352_cast_fp16 = transpose(perm = transpose_352_perm_0, x = var_3529_cast_fp16_0)[name = string("transpose_3695")]; tensor reshape_528_cast_fp16 = reshape(shape = concat_1764, x = transpose_352_cast_fp16)[name = string("reshape_528_cast_fp16")]; tensor transpose_353_perm_0 = const()[name = string("transpose_353_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1765 = const()[name = string("concat_1765"), val = tensor([1, 64, 104])]; tensor transpose_353_cast_fp16 = transpose(perm = transpose_353_perm_0, x = var_3546_cast_fp16_0)[name = string("transpose_3694")]; tensor reshape_529_cast_fp16 = reshape(shape = concat_1765, x = transpose_353_cast_fp16)[name = string("reshape_529_cast_fp16")]; bool matmul_176_transpose_x_0 = const()[name = string("matmul_176_transpose_x_0"), val = bool(false)]; bool matmul_176_transpose_y_0 = const()[name = string("matmul_176_transpose_y_0"), val = bool(false)]; tensor matmul_176_cast_fp16 = matmul(transpose_x = matmul_176_transpose_x_0, transpose_y = matmul_176_transpose_y_0, x = reshape_528_cast_fp16, y = reshape_529_cast_fp16)[name = string("matmul_176_cast_fp16")]; tensor concat_1769 = const()[name = string("concat_1769"), val = tensor([1, 1, 104, 104])]; tensor reshape_530_cast_fp16 = reshape(shape = concat_1769, x = matmul_176_cast_fp16)[name = string("reshape_530_cast_fp16")]; tensor transpose_2864_perm_0 = const()[name = string("transpose_2864_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2864 = transpose(perm = transpose_2864_perm_0, x = reshape_530_cast_fp16)[name = string("transpose_3693")]; tensor w_707_cast_fp16 = add(x = transpose_2864, y = transpose_2305)[name = string("w_707_cast_fp16")]; tensor var_3585_cast_fp16 = softmax(axis = var_3473, x = w_707_cast_fp16)[name = string("op_3585_cast_fp16")]; string var_3587_equation_0 = const()[name = string("op_3587_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3587_cast_fp16 = einsum(equation = var_3587_equation_0, values = (var_3563_cast_fp16_0, var_3585_cast_fp16))[name = string("op_3587_cast_fp16")]; tensor transpose_354_perm_0 = const()[name = string("transpose_354_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1774 = const()[name = string("concat_1774"), val = tensor([1, 104, 64])]; tensor transpose_354_cast_fp16 = transpose(perm = transpose_354_perm_0, x = var_3529_cast_fp16_1)[name = string("transpose_3692")]; tensor reshape_531_cast_fp16 = reshape(shape = concat_1774, x = transpose_354_cast_fp16)[name = string("reshape_531_cast_fp16")]; tensor transpose_355_perm_0 = const()[name = string("transpose_355_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1775 = const()[name = string("concat_1775"), val = tensor([1, 64, 104])]; tensor transpose_355_cast_fp16 = transpose(perm = transpose_355_perm_0, x = var_3546_cast_fp16_1)[name = string("transpose_3691")]; tensor reshape_532_cast_fp16 = reshape(shape = concat_1775, x = transpose_355_cast_fp16)[name = string("reshape_532_cast_fp16")]; bool matmul_177_transpose_x_0 = const()[name = string("matmul_177_transpose_x_0"), val = bool(false)]; bool matmul_177_transpose_y_0 = const()[name = string("matmul_177_transpose_y_0"), val = bool(false)]; tensor matmul_177_cast_fp16 = matmul(transpose_x = matmul_177_transpose_x_0, transpose_y = matmul_177_transpose_y_0, x = reshape_531_cast_fp16, y = reshape_532_cast_fp16)[name = string("matmul_177_cast_fp16")]; tensor concat_1779 = const()[name = string("concat_1779"), val = tensor([1, 1, 104, 104])]; tensor reshape_533_cast_fp16 = reshape(shape = concat_1779, x = matmul_177_cast_fp16)[name = string("reshape_533_cast_fp16")]; tensor transpose_2865_perm_0 = const()[name = string("transpose_2865_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2865 = transpose(perm = transpose_2865_perm_0, x = reshape_533_cast_fp16)[name = string("transpose_3690")]; tensor w_711_cast_fp16 = add(x = transpose_2865, y = transpose_2305)[name = string("w_711_cast_fp16")]; tensor var_3593_cast_fp16 = softmax(axis = var_3473, x = w_711_cast_fp16)[name = string("op_3593_cast_fp16")]; string var_3595_equation_0 = const()[name = string("op_3595_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3595_cast_fp16 = einsum(equation = var_3595_equation_0, values = (var_3563_cast_fp16_1, var_3593_cast_fp16))[name = string("op_3595_cast_fp16")]; tensor transpose_356_perm_0 = const()[name = string("transpose_356_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1784 = const()[name = string("concat_1784"), val = tensor([1, 104, 64])]; tensor transpose_356_cast_fp16 = transpose(perm = transpose_356_perm_0, x = var_3529_cast_fp16_2)[name = string("transpose_3689")]; tensor reshape_534_cast_fp16 = reshape(shape = concat_1784, x = transpose_356_cast_fp16)[name = string("reshape_534_cast_fp16")]; tensor transpose_357_perm_0 = const()[name = string("transpose_357_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1785 = const()[name = string("concat_1785"), val = tensor([1, 64, 104])]; tensor transpose_357_cast_fp16 = transpose(perm = transpose_357_perm_0, x = var_3546_cast_fp16_2)[name = string("transpose_3688")]; tensor reshape_535_cast_fp16 = reshape(shape = concat_1785, x = transpose_357_cast_fp16)[name = string("reshape_535_cast_fp16")]; bool matmul_178_transpose_x_0 = const()[name = string("matmul_178_transpose_x_0"), val = bool(false)]; bool matmul_178_transpose_y_0 = const()[name = string("matmul_178_transpose_y_0"), val = bool(false)]; tensor matmul_178_cast_fp16 = matmul(transpose_x = matmul_178_transpose_x_0, transpose_y = matmul_178_transpose_y_0, x = reshape_534_cast_fp16, y = reshape_535_cast_fp16)[name = string("matmul_178_cast_fp16")]; tensor concat_1789 = const()[name = string("concat_1789"), val = tensor([1, 1, 104, 104])]; tensor reshape_536_cast_fp16 = reshape(shape = concat_1789, x = matmul_178_cast_fp16)[name = string("reshape_536_cast_fp16")]; tensor transpose_2866_perm_0 = const()[name = string("transpose_2866_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2866 = transpose(perm = transpose_2866_perm_0, x = reshape_536_cast_fp16)[name = string("transpose_3687")]; tensor w_715_cast_fp16 = add(x = transpose_2866, y = transpose_2305)[name = string("w_715_cast_fp16")]; tensor var_3601_cast_fp16 = softmax(axis = var_3473, x = w_715_cast_fp16)[name = string("op_3601_cast_fp16")]; string var_3603_equation_0 = const()[name = string("op_3603_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3603_cast_fp16 = einsum(equation = var_3603_equation_0, values = (var_3563_cast_fp16_2, var_3601_cast_fp16))[name = string("op_3603_cast_fp16")]; tensor transpose_358_perm_0 = const()[name = string("transpose_358_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1794 = const()[name = string("concat_1794"), val = tensor([1, 104, 64])]; tensor transpose_358_cast_fp16 = transpose(perm = transpose_358_perm_0, x = var_3529_cast_fp16_3)[name = string("transpose_3686")]; tensor reshape_537_cast_fp16 = reshape(shape = concat_1794, x = transpose_358_cast_fp16)[name = string("reshape_537_cast_fp16")]; tensor transpose_359_perm_0 = const()[name = string("transpose_359_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1795 = const()[name = string("concat_1795"), val = tensor([1, 64, 104])]; tensor transpose_359_cast_fp16 = transpose(perm = transpose_359_perm_0, x = var_3546_cast_fp16_3)[name = string("transpose_3685")]; tensor reshape_538_cast_fp16 = reshape(shape = concat_1795, x = transpose_359_cast_fp16)[name = string("reshape_538_cast_fp16")]; bool matmul_179_transpose_x_0 = const()[name = string("matmul_179_transpose_x_0"), val = bool(false)]; bool matmul_179_transpose_y_0 = const()[name = string("matmul_179_transpose_y_0"), val = bool(false)]; tensor matmul_179_cast_fp16 = matmul(transpose_x = matmul_179_transpose_x_0, transpose_y = matmul_179_transpose_y_0, x = reshape_537_cast_fp16, y = reshape_538_cast_fp16)[name = string("matmul_179_cast_fp16")]; tensor concat_1799 = const()[name = string("concat_1799"), val = tensor([1, 1, 104, 104])]; tensor reshape_539_cast_fp16 = reshape(shape = concat_1799, x = matmul_179_cast_fp16)[name = string("reshape_539_cast_fp16")]; tensor transpose_2867_perm_0 = const()[name = string("transpose_2867_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2867 = transpose(perm = transpose_2867_perm_0, x = reshape_539_cast_fp16)[name = string("transpose_3684")]; tensor w_719_cast_fp16 = add(x = transpose_2867, y = transpose_2305)[name = string("w_719_cast_fp16")]; tensor var_3609_cast_fp16 = softmax(axis = var_3473, x = w_719_cast_fp16)[name = string("op_3609_cast_fp16")]; string var_3611_equation_0 = const()[name = string("op_3611_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3611_cast_fp16 = einsum(equation = var_3611_equation_0, values = (var_3563_cast_fp16_3, var_3609_cast_fp16))[name = string("op_3611_cast_fp16")]; tensor transpose_360_perm_0 = const()[name = string("transpose_360_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1804 = const()[name = string("concat_1804"), val = tensor([1, 104, 64])]; tensor transpose_360_cast_fp16 = transpose(perm = transpose_360_perm_0, x = var_3529_cast_fp16_4)[name = string("transpose_3683")]; tensor reshape_540_cast_fp16 = reshape(shape = concat_1804, x = transpose_360_cast_fp16)[name = string("reshape_540_cast_fp16")]; tensor transpose_361_perm_0 = const()[name = string("transpose_361_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1805 = const()[name = string("concat_1805"), val = tensor([1, 64, 104])]; tensor transpose_361_cast_fp16 = transpose(perm = transpose_361_perm_0, x = var_3546_cast_fp16_4)[name = string("transpose_3682")]; tensor reshape_541_cast_fp16 = reshape(shape = concat_1805, x = transpose_361_cast_fp16)[name = string("reshape_541_cast_fp16")]; bool matmul_180_transpose_x_0 = const()[name = string("matmul_180_transpose_x_0"), val = bool(false)]; bool matmul_180_transpose_y_0 = const()[name = string("matmul_180_transpose_y_0"), val = bool(false)]; tensor matmul_180_cast_fp16 = matmul(transpose_x = matmul_180_transpose_x_0, transpose_y = matmul_180_transpose_y_0, x = reshape_540_cast_fp16, y = reshape_541_cast_fp16)[name = string("matmul_180_cast_fp16")]; tensor concat_1809 = const()[name = string("concat_1809"), val = tensor([1, 1, 104, 104])]; tensor reshape_542_cast_fp16 = reshape(shape = concat_1809, x = matmul_180_cast_fp16)[name = string("reshape_542_cast_fp16")]; tensor transpose_2868_perm_0 = const()[name = string("transpose_2868_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2868 = transpose(perm = transpose_2868_perm_0, x = reshape_542_cast_fp16)[name = string("transpose_3681")]; tensor w_723_cast_fp16 = add(x = transpose_2868, y = transpose_2305)[name = string("w_723_cast_fp16")]; tensor var_3617_cast_fp16 = softmax(axis = var_3473, x = w_723_cast_fp16)[name = string("op_3617_cast_fp16")]; string var_3619_equation_0 = const()[name = string("op_3619_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3619_cast_fp16 = einsum(equation = var_3619_equation_0, values = (var_3563_cast_fp16_4, var_3617_cast_fp16))[name = string("op_3619_cast_fp16")]; tensor transpose_362_perm_0 = const()[name = string("transpose_362_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1814 = const()[name = string("concat_1814"), val = tensor([1, 104, 64])]; tensor transpose_362_cast_fp16 = transpose(perm = transpose_362_perm_0, x = var_3529_cast_fp16_5)[name = string("transpose_3680")]; tensor reshape_543_cast_fp16 = reshape(shape = concat_1814, x = transpose_362_cast_fp16)[name = string("reshape_543_cast_fp16")]; tensor transpose_363_perm_0 = const()[name = string("transpose_363_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1815 = const()[name = string("concat_1815"), val = tensor([1, 64, 104])]; tensor transpose_363_cast_fp16 = transpose(perm = transpose_363_perm_0, x = var_3546_cast_fp16_5)[name = string("transpose_3679")]; tensor reshape_544_cast_fp16 = reshape(shape = concat_1815, x = transpose_363_cast_fp16)[name = string("reshape_544_cast_fp16")]; bool matmul_181_transpose_x_0 = const()[name = string("matmul_181_transpose_x_0"), val = bool(false)]; bool matmul_181_transpose_y_0 = const()[name = string("matmul_181_transpose_y_0"), val = bool(false)]; tensor matmul_181_cast_fp16 = matmul(transpose_x = matmul_181_transpose_x_0, transpose_y = matmul_181_transpose_y_0, x = reshape_543_cast_fp16, y = reshape_544_cast_fp16)[name = string("matmul_181_cast_fp16")]; tensor concat_1819 = const()[name = string("concat_1819"), val = tensor([1, 1, 104, 104])]; tensor reshape_545_cast_fp16 = reshape(shape = concat_1819, x = matmul_181_cast_fp16)[name = string("reshape_545_cast_fp16")]; tensor transpose_2869_perm_0 = const()[name = string("transpose_2869_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2869 = transpose(perm = transpose_2869_perm_0, x = reshape_545_cast_fp16)[name = string("transpose_3678")]; tensor w_727_cast_fp16 = add(x = transpose_2869, y = transpose_2305)[name = string("w_727_cast_fp16")]; tensor var_3625_cast_fp16 = softmax(axis = var_3473, x = w_727_cast_fp16)[name = string("op_3625_cast_fp16")]; string var_3627_equation_0 = const()[name = string("op_3627_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3627_cast_fp16 = einsum(equation = var_3627_equation_0, values = (var_3563_cast_fp16_5, var_3625_cast_fp16))[name = string("op_3627_cast_fp16")]; tensor transpose_364_perm_0 = const()[name = string("transpose_364_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1824 = const()[name = string("concat_1824"), val = tensor([1, 104, 64])]; tensor transpose_364_cast_fp16 = transpose(perm = transpose_364_perm_0, x = var_3529_cast_fp16_6)[name = string("transpose_3677")]; tensor reshape_546_cast_fp16 = reshape(shape = concat_1824, x = transpose_364_cast_fp16)[name = string("reshape_546_cast_fp16")]; tensor transpose_365_perm_0 = const()[name = string("transpose_365_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1825 = const()[name = string("concat_1825"), val = tensor([1, 64, 104])]; tensor transpose_365_cast_fp16 = transpose(perm = transpose_365_perm_0, x = var_3546_cast_fp16_6)[name = string("transpose_3676")]; tensor reshape_547_cast_fp16 = reshape(shape = concat_1825, x = transpose_365_cast_fp16)[name = string("reshape_547_cast_fp16")]; bool matmul_182_transpose_x_0 = const()[name = string("matmul_182_transpose_x_0"), val = bool(false)]; bool matmul_182_transpose_y_0 = const()[name = string("matmul_182_transpose_y_0"), val = bool(false)]; tensor matmul_182_cast_fp16 = matmul(transpose_x = matmul_182_transpose_x_0, transpose_y = matmul_182_transpose_y_0, x = reshape_546_cast_fp16, y = reshape_547_cast_fp16)[name = string("matmul_182_cast_fp16")]; tensor concat_1829 = const()[name = string("concat_1829"), val = tensor([1, 1, 104, 104])]; tensor reshape_548_cast_fp16 = reshape(shape = concat_1829, x = matmul_182_cast_fp16)[name = string("reshape_548_cast_fp16")]; tensor transpose_2870_perm_0 = const()[name = string("transpose_2870_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2870 = transpose(perm = transpose_2870_perm_0, x = reshape_548_cast_fp16)[name = string("transpose_3675")]; tensor w_731_cast_fp16 = add(x = transpose_2870, y = transpose_2305)[name = string("w_731_cast_fp16")]; tensor var_3633_cast_fp16 = softmax(axis = var_3473, x = w_731_cast_fp16)[name = string("op_3633_cast_fp16")]; string var_3635_equation_0 = const()[name = string("op_3635_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3635_cast_fp16 = einsum(equation = var_3635_equation_0, values = (var_3563_cast_fp16_6, var_3633_cast_fp16))[name = string("op_3635_cast_fp16")]; tensor transpose_366_perm_0 = const()[name = string("transpose_366_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1834 = const()[name = string("concat_1834"), val = tensor([1, 104, 64])]; tensor transpose_366_cast_fp16 = transpose(perm = transpose_366_perm_0, x = var_3529_cast_fp16_7)[name = string("transpose_3674")]; tensor reshape_549_cast_fp16 = reshape(shape = concat_1834, x = transpose_366_cast_fp16)[name = string("reshape_549_cast_fp16")]; tensor transpose_367_perm_0 = const()[name = string("transpose_367_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1835 = const()[name = string("concat_1835"), val = tensor([1, 64, 104])]; tensor transpose_367_cast_fp16 = transpose(perm = transpose_367_perm_0, x = var_3546_cast_fp16_7)[name = string("transpose_3673")]; tensor reshape_550_cast_fp16 = reshape(shape = concat_1835, x = transpose_367_cast_fp16)[name = string("reshape_550_cast_fp16")]; bool matmul_183_transpose_x_0 = const()[name = string("matmul_183_transpose_x_0"), val = bool(false)]; bool matmul_183_transpose_y_0 = const()[name = string("matmul_183_transpose_y_0"), val = bool(false)]; tensor matmul_183_cast_fp16 = matmul(transpose_x = matmul_183_transpose_x_0, transpose_y = matmul_183_transpose_y_0, x = reshape_549_cast_fp16, y = reshape_550_cast_fp16)[name = string("matmul_183_cast_fp16")]; tensor concat_1839 = const()[name = string("concat_1839"), val = tensor([1, 1, 104, 104])]; tensor reshape_551_cast_fp16 = reshape(shape = concat_1839, x = matmul_183_cast_fp16)[name = string("reshape_551_cast_fp16")]; tensor transpose_2871_perm_0 = const()[name = string("transpose_2871_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2871 = transpose(perm = transpose_2871_perm_0, x = reshape_551_cast_fp16)[name = string("transpose_3672")]; tensor w_735_cast_fp16 = add(x = transpose_2871, y = transpose_2305)[name = string("w_735_cast_fp16")]; tensor var_3641_cast_fp16 = softmax(axis = var_3473, x = w_735_cast_fp16)[name = string("op_3641_cast_fp16")]; string var_3643_equation_0 = const()[name = string("op_3643_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3643_cast_fp16 = einsum(equation = var_3643_equation_0, values = (var_3563_cast_fp16_7, var_3641_cast_fp16))[name = string("op_3643_cast_fp16")]; tensor transpose_368_perm_0 = const()[name = string("transpose_368_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1844 = const()[name = string("concat_1844"), val = tensor([1, 104, 64])]; tensor transpose_368_cast_fp16 = transpose(perm = transpose_368_perm_0, x = var_3529_cast_fp16_8)[name = string("transpose_3671")]; tensor reshape_552_cast_fp16 = reshape(shape = concat_1844, x = transpose_368_cast_fp16)[name = string("reshape_552_cast_fp16")]; tensor transpose_369_perm_0 = const()[name = string("transpose_369_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1845 = const()[name = string("concat_1845"), val = tensor([1, 64, 104])]; tensor transpose_369_cast_fp16 = transpose(perm = transpose_369_perm_0, x = var_3546_cast_fp16_8)[name = string("transpose_3670")]; tensor reshape_553_cast_fp16 = reshape(shape = concat_1845, x = transpose_369_cast_fp16)[name = string("reshape_553_cast_fp16")]; bool matmul_184_transpose_x_0 = const()[name = string("matmul_184_transpose_x_0"), val = bool(false)]; bool matmul_184_transpose_y_0 = const()[name = string("matmul_184_transpose_y_0"), val = bool(false)]; tensor matmul_184_cast_fp16 = matmul(transpose_x = matmul_184_transpose_x_0, transpose_y = matmul_184_transpose_y_0, x = reshape_552_cast_fp16, y = reshape_553_cast_fp16)[name = string("matmul_184_cast_fp16")]; tensor concat_1849 = const()[name = string("concat_1849"), val = tensor([1, 1, 104, 104])]; tensor reshape_554_cast_fp16 = reshape(shape = concat_1849, x = matmul_184_cast_fp16)[name = string("reshape_554_cast_fp16")]; tensor transpose_2872_perm_0 = const()[name = string("transpose_2872_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2872 = transpose(perm = transpose_2872_perm_0, x = reshape_554_cast_fp16)[name = string("transpose_3669")]; tensor w_739_cast_fp16 = add(x = transpose_2872, y = transpose_2305)[name = string("w_739_cast_fp16")]; tensor var_3649_cast_fp16 = softmax(axis = var_3473, x = w_739_cast_fp16)[name = string("op_3649_cast_fp16")]; string var_3651_equation_0 = const()[name = string("op_3651_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3651_cast_fp16 = einsum(equation = var_3651_equation_0, values = (var_3563_cast_fp16_8, var_3649_cast_fp16))[name = string("op_3651_cast_fp16")]; tensor transpose_370_perm_0 = const()[name = string("transpose_370_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1854 = const()[name = string("concat_1854"), val = tensor([1, 104, 64])]; tensor transpose_370_cast_fp16 = transpose(perm = transpose_370_perm_0, x = var_3529_cast_fp16_9)[name = string("transpose_3668")]; tensor reshape_555_cast_fp16 = reshape(shape = concat_1854, x = transpose_370_cast_fp16)[name = string("reshape_555_cast_fp16")]; tensor transpose_371_perm_0 = const()[name = string("transpose_371_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1855 = const()[name = string("concat_1855"), val = tensor([1, 64, 104])]; tensor transpose_371_cast_fp16 = transpose(perm = transpose_371_perm_0, x = var_3546_cast_fp16_9)[name = string("transpose_3667")]; tensor reshape_556_cast_fp16 = reshape(shape = concat_1855, x = transpose_371_cast_fp16)[name = string("reshape_556_cast_fp16")]; bool matmul_185_transpose_x_0 = const()[name = string("matmul_185_transpose_x_0"), val = bool(false)]; bool matmul_185_transpose_y_0 = const()[name = string("matmul_185_transpose_y_0"), val = bool(false)]; tensor matmul_185_cast_fp16 = matmul(transpose_x = matmul_185_transpose_x_0, transpose_y = matmul_185_transpose_y_0, x = reshape_555_cast_fp16, y = reshape_556_cast_fp16)[name = string("matmul_185_cast_fp16")]; tensor concat_1859 = const()[name = string("concat_1859"), val = tensor([1, 1, 104, 104])]; tensor reshape_557_cast_fp16 = reshape(shape = concat_1859, x = matmul_185_cast_fp16)[name = string("reshape_557_cast_fp16")]; tensor transpose_2873_perm_0 = const()[name = string("transpose_2873_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2873 = transpose(perm = transpose_2873_perm_0, x = reshape_557_cast_fp16)[name = string("transpose_3666")]; tensor w_743_cast_fp16 = add(x = transpose_2873, y = transpose_2305)[name = string("w_743_cast_fp16")]; tensor var_3657_cast_fp16 = softmax(axis = var_3473, x = w_743_cast_fp16)[name = string("op_3657_cast_fp16")]; string var_3659_equation_0 = const()[name = string("op_3659_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3659_cast_fp16 = einsum(equation = var_3659_equation_0, values = (var_3563_cast_fp16_9, var_3657_cast_fp16))[name = string("op_3659_cast_fp16")]; tensor transpose_372_perm_0 = const()[name = string("transpose_372_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1864 = const()[name = string("concat_1864"), val = tensor([1, 104, 64])]; tensor transpose_372_cast_fp16 = transpose(perm = transpose_372_perm_0, x = var_3529_cast_fp16_10)[name = string("transpose_3665")]; tensor reshape_558_cast_fp16 = reshape(shape = concat_1864, x = transpose_372_cast_fp16)[name = string("reshape_558_cast_fp16")]; tensor transpose_373_perm_0 = const()[name = string("transpose_373_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1865 = const()[name = string("concat_1865"), val = tensor([1, 64, 104])]; tensor transpose_373_cast_fp16 = transpose(perm = transpose_373_perm_0, x = var_3546_cast_fp16_10)[name = string("transpose_3664")]; tensor reshape_559_cast_fp16 = reshape(shape = concat_1865, x = transpose_373_cast_fp16)[name = string("reshape_559_cast_fp16")]; bool matmul_186_transpose_x_0 = const()[name = string("matmul_186_transpose_x_0"), val = bool(false)]; bool matmul_186_transpose_y_0 = const()[name = string("matmul_186_transpose_y_0"), val = bool(false)]; tensor matmul_186_cast_fp16 = matmul(transpose_x = matmul_186_transpose_x_0, transpose_y = matmul_186_transpose_y_0, x = reshape_558_cast_fp16, y = reshape_559_cast_fp16)[name = string("matmul_186_cast_fp16")]; tensor concat_1869 = const()[name = string("concat_1869"), val = tensor([1, 1, 104, 104])]; tensor reshape_560_cast_fp16 = reshape(shape = concat_1869, x = matmul_186_cast_fp16)[name = string("reshape_560_cast_fp16")]; tensor transpose_2874_perm_0 = const()[name = string("transpose_2874_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2874 = transpose(perm = transpose_2874_perm_0, x = reshape_560_cast_fp16)[name = string("transpose_3663")]; tensor w_747_cast_fp16 = add(x = transpose_2874, y = transpose_2305)[name = string("w_747_cast_fp16")]; tensor var_3665_cast_fp16 = softmax(axis = var_3473, x = w_747_cast_fp16)[name = string("op_3665_cast_fp16")]; string var_3667_equation_0 = const()[name = string("op_3667_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3667_cast_fp16 = einsum(equation = var_3667_equation_0, values = (var_3563_cast_fp16_10, var_3665_cast_fp16))[name = string("op_3667_cast_fp16")]; tensor transpose_374_perm_0 = const()[name = string("transpose_374_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1874 = const()[name = string("concat_1874"), val = tensor([1, 104, 64])]; tensor transpose_374_cast_fp16 = transpose(perm = transpose_374_perm_0, x = var_3529_cast_fp16_11)[name = string("transpose_3662")]; tensor reshape_561_cast_fp16 = reshape(shape = concat_1874, x = transpose_374_cast_fp16)[name = string("reshape_561_cast_fp16")]; tensor transpose_375_perm_0 = const()[name = string("transpose_375_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1875 = const()[name = string("concat_1875"), val = tensor([1, 64, 104])]; tensor transpose_375_cast_fp16 = transpose(perm = transpose_375_perm_0, x = var_3546_cast_fp16_11)[name = string("transpose_3661")]; tensor reshape_562_cast_fp16 = reshape(shape = concat_1875, x = transpose_375_cast_fp16)[name = string("reshape_562_cast_fp16")]; bool matmul_187_transpose_x_0 = const()[name = string("matmul_187_transpose_x_0"), val = bool(false)]; bool matmul_187_transpose_y_0 = const()[name = string("matmul_187_transpose_y_0"), val = bool(false)]; tensor matmul_187_cast_fp16 = matmul(transpose_x = matmul_187_transpose_x_0, transpose_y = matmul_187_transpose_y_0, x = reshape_561_cast_fp16, y = reshape_562_cast_fp16)[name = string("matmul_187_cast_fp16")]; tensor concat_1879 = const()[name = string("concat_1879"), val = tensor([1, 1, 104, 104])]; tensor reshape_563_cast_fp16 = reshape(shape = concat_1879, x = matmul_187_cast_fp16)[name = string("reshape_563_cast_fp16")]; tensor transpose_2875_perm_0 = const()[name = string("transpose_2875_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2875 = transpose(perm = transpose_2875_perm_0, x = reshape_563_cast_fp16)[name = string("transpose_3660")]; tensor w_751_cast_fp16 = add(x = transpose_2875, y = transpose_2305)[name = string("w_751_cast_fp16")]; tensor var_3673_cast_fp16 = softmax(axis = var_3473, x = w_751_cast_fp16)[name = string("op_3673_cast_fp16")]; string var_3675_equation_0 = const()[name = string("op_3675_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3675_cast_fp16 = einsum(equation = var_3675_equation_0, values = (var_3563_cast_fp16_11, var_3673_cast_fp16))[name = string("op_3675_cast_fp16")]; tensor transpose_376_perm_0 = const()[name = string("transpose_376_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1884 = const()[name = string("concat_1884"), val = tensor([1, 104, 64])]; tensor transpose_376_cast_fp16 = transpose(perm = transpose_376_perm_0, x = var_3529_cast_fp16_12)[name = string("transpose_3659")]; tensor reshape_564_cast_fp16 = reshape(shape = concat_1884, x = transpose_376_cast_fp16)[name = string("reshape_564_cast_fp16")]; tensor transpose_377_perm_0 = const()[name = string("transpose_377_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1885 = const()[name = string("concat_1885"), val = tensor([1, 64, 104])]; tensor transpose_377_cast_fp16 = transpose(perm = transpose_377_perm_0, x = var_3546_cast_fp16_12)[name = string("transpose_3658")]; tensor reshape_565_cast_fp16 = reshape(shape = concat_1885, x = transpose_377_cast_fp16)[name = string("reshape_565_cast_fp16")]; bool matmul_188_transpose_x_0 = const()[name = string("matmul_188_transpose_x_0"), val = bool(false)]; bool matmul_188_transpose_y_0 = const()[name = string("matmul_188_transpose_y_0"), val = bool(false)]; tensor matmul_188_cast_fp16 = matmul(transpose_x = matmul_188_transpose_x_0, transpose_y = matmul_188_transpose_y_0, x = reshape_564_cast_fp16, y = reshape_565_cast_fp16)[name = string("matmul_188_cast_fp16")]; tensor concat_1889 = const()[name = string("concat_1889"), val = tensor([1, 1, 104, 104])]; tensor reshape_566_cast_fp16 = reshape(shape = concat_1889, x = matmul_188_cast_fp16)[name = string("reshape_566_cast_fp16")]; tensor transpose_2876_perm_0 = const()[name = string("transpose_2876_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2876 = transpose(perm = transpose_2876_perm_0, x = reshape_566_cast_fp16)[name = string("transpose_3657")]; tensor w_755_cast_fp16 = add(x = transpose_2876, y = transpose_2305)[name = string("w_755_cast_fp16")]; tensor var_3681_cast_fp16 = softmax(axis = var_3473, x = w_755_cast_fp16)[name = string("op_3681_cast_fp16")]; string var_3683_equation_0 = const()[name = string("op_3683_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3683_cast_fp16 = einsum(equation = var_3683_equation_0, values = (var_3563_cast_fp16_12, var_3681_cast_fp16))[name = string("op_3683_cast_fp16")]; tensor transpose_378_perm_0 = const()[name = string("transpose_378_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1894 = const()[name = string("concat_1894"), val = tensor([1, 104, 64])]; tensor transpose_378_cast_fp16 = transpose(perm = transpose_378_perm_0, x = var_3529_cast_fp16_13)[name = string("transpose_3656")]; tensor reshape_567_cast_fp16 = reshape(shape = concat_1894, x = transpose_378_cast_fp16)[name = string("reshape_567_cast_fp16")]; tensor transpose_379_perm_0 = const()[name = string("transpose_379_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1895 = const()[name = string("concat_1895"), val = tensor([1, 64, 104])]; tensor transpose_379_cast_fp16 = transpose(perm = transpose_379_perm_0, x = var_3546_cast_fp16_13)[name = string("transpose_3655")]; tensor reshape_568_cast_fp16 = reshape(shape = concat_1895, x = transpose_379_cast_fp16)[name = string("reshape_568_cast_fp16")]; bool matmul_189_transpose_x_0 = const()[name = string("matmul_189_transpose_x_0"), val = bool(false)]; bool matmul_189_transpose_y_0 = const()[name = string("matmul_189_transpose_y_0"), val = bool(false)]; tensor matmul_189_cast_fp16 = matmul(transpose_x = matmul_189_transpose_x_0, transpose_y = matmul_189_transpose_y_0, x = reshape_567_cast_fp16, y = reshape_568_cast_fp16)[name = string("matmul_189_cast_fp16")]; tensor concat_1899 = const()[name = string("concat_1899"), val = tensor([1, 1, 104, 104])]; tensor reshape_569_cast_fp16 = reshape(shape = concat_1899, x = matmul_189_cast_fp16)[name = string("reshape_569_cast_fp16")]; tensor transpose_2877_perm_0 = const()[name = string("transpose_2877_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2877 = transpose(perm = transpose_2877_perm_0, x = reshape_569_cast_fp16)[name = string("transpose_3654")]; tensor w_759_cast_fp16 = add(x = transpose_2877, y = transpose_2305)[name = string("w_759_cast_fp16")]; tensor var_3689_cast_fp16 = softmax(axis = var_3473, x = w_759_cast_fp16)[name = string("op_3689_cast_fp16")]; string var_3691_equation_0 = const()[name = string("op_3691_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3691_cast_fp16 = einsum(equation = var_3691_equation_0, values = (var_3563_cast_fp16_13, var_3689_cast_fp16))[name = string("op_3691_cast_fp16")]; tensor transpose_380_perm_0 = const()[name = string("transpose_380_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1904 = const()[name = string("concat_1904"), val = tensor([1, 104, 64])]; tensor transpose_380_cast_fp16 = transpose(perm = transpose_380_perm_0, x = var_3529_cast_fp16_14)[name = string("transpose_3653")]; tensor reshape_570_cast_fp16 = reshape(shape = concat_1904, x = transpose_380_cast_fp16)[name = string("reshape_570_cast_fp16")]; tensor transpose_381_perm_0 = const()[name = string("transpose_381_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1905 = const()[name = string("concat_1905"), val = tensor([1, 64, 104])]; tensor transpose_381_cast_fp16 = transpose(perm = transpose_381_perm_0, x = var_3546_cast_fp16_14)[name = string("transpose_3652")]; tensor reshape_571_cast_fp16 = reshape(shape = concat_1905, x = transpose_381_cast_fp16)[name = string("reshape_571_cast_fp16")]; bool matmul_190_transpose_x_0 = const()[name = string("matmul_190_transpose_x_0"), val = bool(false)]; bool matmul_190_transpose_y_0 = const()[name = string("matmul_190_transpose_y_0"), val = bool(false)]; tensor matmul_190_cast_fp16 = matmul(transpose_x = matmul_190_transpose_x_0, transpose_y = matmul_190_transpose_y_0, x = reshape_570_cast_fp16, y = reshape_571_cast_fp16)[name = string("matmul_190_cast_fp16")]; tensor concat_1909 = const()[name = string("concat_1909"), val = tensor([1, 1, 104, 104])]; tensor reshape_572_cast_fp16 = reshape(shape = concat_1909, x = matmul_190_cast_fp16)[name = string("reshape_572_cast_fp16")]; tensor transpose_2878_perm_0 = const()[name = string("transpose_2878_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2878 = transpose(perm = transpose_2878_perm_0, x = reshape_572_cast_fp16)[name = string("transpose_3651")]; tensor w_763_cast_fp16 = add(x = transpose_2878, y = transpose_2305)[name = string("w_763_cast_fp16")]; tensor var_3697_cast_fp16 = softmax(axis = var_3473, x = w_763_cast_fp16)[name = string("op_3697_cast_fp16")]; string var_3699_equation_0 = const()[name = string("op_3699_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3699_cast_fp16 = einsum(equation = var_3699_equation_0, values = (var_3563_cast_fp16_14, var_3697_cast_fp16))[name = string("op_3699_cast_fp16")]; tensor transpose_382_perm_0 = const()[name = string("transpose_382_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1914 = const()[name = string("concat_1914"), val = tensor([1, 104, 64])]; tensor transpose_382_cast_fp16 = transpose(perm = transpose_382_perm_0, x = var_3529_cast_fp16_15)[name = string("transpose_3650")]; tensor reshape_573_cast_fp16 = reshape(shape = concat_1914, x = transpose_382_cast_fp16)[name = string("reshape_573_cast_fp16")]; tensor transpose_383_perm_0 = const()[name = string("transpose_383_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1915 = const()[name = string("concat_1915"), val = tensor([1, 64, 104])]; tensor transpose_383_cast_fp16 = transpose(perm = transpose_383_perm_0, x = var_3546_cast_fp16_15)[name = string("transpose_3649")]; tensor reshape_574_cast_fp16 = reshape(shape = concat_1915, x = transpose_383_cast_fp16)[name = string("reshape_574_cast_fp16")]; bool matmul_191_transpose_x_0 = const()[name = string("matmul_191_transpose_x_0"), val = bool(false)]; bool matmul_191_transpose_y_0 = const()[name = string("matmul_191_transpose_y_0"), val = bool(false)]; tensor matmul_191_cast_fp16 = matmul(transpose_x = matmul_191_transpose_x_0, transpose_y = matmul_191_transpose_y_0, x = reshape_573_cast_fp16, y = reshape_574_cast_fp16)[name = string("matmul_191_cast_fp16")]; tensor concat_1919 = const()[name = string("concat_1919"), val = tensor([1, 1, 104, 104])]; tensor reshape_575_cast_fp16 = reshape(shape = concat_1919, x = matmul_191_cast_fp16)[name = string("reshape_575_cast_fp16")]; tensor transpose_2879_perm_0 = const()[name = string("transpose_2879_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2879 = transpose(perm = transpose_2879_perm_0, x = reshape_575_cast_fp16)[name = string("transpose_3648")]; tensor w_767_cast_fp16 = add(x = transpose_2879, y = transpose_2305)[name = string("w_767_cast_fp16")]; tensor var_3705_cast_fp16 = softmax(axis = var_3473, x = w_767_cast_fp16)[name = string("op_3705_cast_fp16")]; string var_3707_equation_0 = const()[name = string("op_3707_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3707_cast_fp16 = einsum(equation = var_3707_equation_0, values = (var_3563_cast_fp16_15, var_3705_cast_fp16))[name = string("op_3707_cast_fp16")]; bool input_99_interleave_0 = const()[name = string("input_99_interleave_0"), val = bool(false)]; tensor input_99_cast_fp16 = concat(axis = var_3473, interleave = input_99_interleave_0, values = (var_3587_cast_fp16, var_3595_cast_fp16, var_3603_cast_fp16, var_3611_cast_fp16, var_3619_cast_fp16, var_3627_cast_fp16, var_3635_cast_fp16, var_3643_cast_fp16, var_3651_cast_fp16, var_3659_cast_fp16, var_3667_cast_fp16, var_3675_cast_fp16, var_3683_cast_fp16, var_3691_cast_fp16, var_3699_cast_fp16, var_3707_cast_fp16))[name = string("input_99_cast_fp16")]; string var_3716_pad_type_0 = const()[name = string("op_3716_pad_type_0"), val = string("valid")]; tensor var_3716_strides_0 = const()[name = string("op_3716_strides_0"), val = tensor([1, 1])]; tensor var_3716_pad_0 = const()[name = string("op_3716_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3716_dilations_0 = const()[name = string("op_3716_dilations_0"), val = tensor([1, 1])]; int32 var_3716_groups_0 = const()[name = string("op_3716_groups_0"), val = int32(1)]; tensor layers_11_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_11_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(307500992)))]; tensor layers_11_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_11_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(309598208)))]; tensor var_3716_cast_fp16 = conv(bias = layers_11_self_attn_out_proj_bias_to_fp16, dilations = var_3716_dilations_0, groups = var_3716_groups_0, pad = var_3716_pad_0, pad_type = var_3716_pad_type_0, strides = var_3716_strides_0, weight = layers_11_self_attn_out_proj_weight_to_fp16, x = input_99_cast_fp16)[name = string("op_3716_cast_fp16")]; tensor x_127_cast_fp16 = add(x = x_123_cast_fp16, y = var_3716_cast_fp16)[name = string("x_127_cast_fp16")]; tensor mu_47_axes_0 = const()[name = string("mu_47_axes_0"), val = tensor([1])]; bool mu_47_keep_dims_0 = const()[name = string("mu_47_keep_dims_0"), val = bool(true)]; tensor mu_47_cast_fp16 = reduce_mean(axes = mu_47_axes_0, keep_dims = mu_47_keep_dims_0, x = x_127_cast_fp16)[name = string("mu_47_cast_fp16")]; tensor var_3722_cast_fp16 = sub(x = x_127_cast_fp16, y = mu_47_cast_fp16)[name = string("op_3722_cast_fp16")]; fp16 var_3476_promoted_1_to_fp16 = const()[name = string("op_3476_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_3723_cast_fp16 = pow(x = var_3722_cast_fp16, y = var_3476_promoted_1_to_fp16)[name = string("op_3723_cast_fp16")]; tensor var_47_axes_0 = const()[name = string("var_47_axes_0"), val = tensor([1])]; bool var_47_keep_dims_0 = const()[name = string("var_47_keep_dims_0"), val = bool(true)]; tensor var_47_cast_fp16 = reduce_mean(axes = var_47_axes_0, keep_dims = var_47_keep_dims_0, x = var_3723_cast_fp16)[name = string("var_47_cast_fp16")]; fp16 var_3727_to_fp16 = const()[name = string("op_3727_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_3728_cast_fp16 = add(x = var_47_cast_fp16, y = var_3727_to_fp16)[name = string("op_3728_cast_fp16")]; fp32 var_3729_epsilon_0 = const()[name = string("op_3729_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_3729_cast_fp16 = rsqrt(epsilon = var_3729_epsilon_0, x = var_3728_cast_fp16)[name = string("op_3729_cast_fp16")]; tensor x_129_cast_fp16 = mul(x = var_3722_cast_fp16, y = var_3729_cast_fp16)[name = string("x_129_cast_fp16")]; tensor input_101_gamma_0_to_fp16 = const()[name = string("input_101_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(309600320)))]; tensor input_101_beta_0_to_fp16 = const()[name = string("input_101_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(309602432)))]; fp16 input_101_epsilon_0_to_fp16 = const()[name = string("input_101_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_101_cast_fp16 = batch_norm(beta = input_101_beta_0_to_fp16, epsilon = input_101_epsilon_0_to_fp16, gamma = input_101_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_129_cast_fp16)[name = string("input_101_cast_fp16")]; string x_131_pad_type_0 = const()[name = string("x_131_pad_type_0"), val = string("valid")]; tensor x_131_strides_0 = const()[name = string("x_131_strides_0"), val = tensor([1, 1])]; tensor x_131_pad_0 = const()[name = string("x_131_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_131_dilations_0 = const()[name = string("x_131_dilations_0"), val = tensor([1, 1])]; int32 x_131_groups_0 = const()[name = string("x_131_groups_0"), val = int32(1)]; tensor layers_11_fc1_weight_to_fp16 = const()[name = string("layers_11_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(309604544)))]; tensor layers_11_fc1_bias_to_fp16 = const()[name = string("layers_11_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(317993216)))]; tensor x_131_cast_fp16 = conv(bias = layers_11_fc1_bias_to_fp16, dilations = x_131_dilations_0, groups = x_131_groups_0, pad = x_131_pad_0, pad_type = x_131_pad_type_0, strides = x_131_strides_0, weight = layers_11_fc1_weight_to_fp16, x = input_101_cast_fp16)[name = string("x_131_cast_fp16")]; fp16 var_3744_to_fp16 = const()[name = string("op_3744_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_3745_cast_fp16 = mul(x = x_131_cast_fp16, y = var_3744_to_fp16)[name = string("op_3745_cast_fp16")]; tensor var_3746_cast_fp16 = mul(x = var_3745_cast_fp16, y = x_131_cast_fp16)[name = string("op_3746_cast_fp16")]; tensor var_3747_cast_fp16 = mul(x = var_3746_cast_fp16, y = x_131_cast_fp16)[name = string("op_3747_cast_fp16")]; tensor var_3748_cast_fp16 = add(x = x_131_cast_fp16, y = var_3747_cast_fp16)[name = string("op_3748_cast_fp16")]; fp16 var_3749_to_fp16 = const()[name = string("op_3749_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_29_cast_fp16 = mul(x = var_3748_cast_fp16, y = var_3749_to_fp16)[name = string("u_29_cast_fp16")]; fp16 var_3751_to_fp16 = const()[name = string("op_3751_to_fp16"), val = fp16(0x1p-1)]; tensor var_3752_cast_fp16 = mul(x = x_131_cast_fp16, y = var_3751_to_fp16)[name = string("op_3752_cast_fp16")]; tensor var_3753_cast_fp16 = tanh(x = u_29_cast_fp16)[name = string("op_3753_cast_fp16")]; fp16 var_3754_to_fp16 = const()[name = string("op_3754_to_fp16"), val = fp16(0x1p+0)]; tensor var_3755_cast_fp16 = add(x = var_3753_cast_fp16, y = var_3754_to_fp16)[name = string("op_3755_cast_fp16")]; tensor input_103_cast_fp16 = mul(x = var_3752_cast_fp16, y = var_3755_cast_fp16)[name = string("input_103_cast_fp16")]; string h_23_pad_type_0 = const()[name = string("h_23_pad_type_0"), val = string("valid")]; tensor h_23_strides_0 = const()[name = string("h_23_strides_0"), val = tensor([1, 1])]; tensor h_23_pad_0 = const()[name = string("h_23_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_23_dilations_0 = const()[name = string("h_23_dilations_0"), val = tensor([1, 1])]; int32 h_23_groups_0 = const()[name = string("h_23_groups_0"), val = int32(1)]; tensor layers_11_fc2_weight_to_fp16 = const()[name = string("layers_11_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(318001472)))]; tensor layers_11_fc2_bias_to_fp16 = const()[name = string("layers_11_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(326390144)))]; tensor h_23_cast_fp16 = conv(bias = layers_11_fc2_bias_to_fp16, dilations = h_23_dilations_0, groups = h_23_groups_0, pad = h_23_pad_0, pad_type = h_23_pad_type_0, strides = h_23_strides_0, weight = layers_11_fc2_weight_to_fp16, x = input_103_cast_fp16)[name = string("h_23_cast_fp16")]; tensor x_133_cast_fp16 = add(x = x_127_cast_fp16, y = h_23_cast_fp16)[name = string("x_133_cast_fp16")]; int32 var_3771 = const()[name = string("op_3771"), val = int32(1)]; tensor mu_49_axes_0 = const()[name = string("mu_49_axes_0"), val = tensor([1])]; bool mu_49_keep_dims_0 = const()[name = string("mu_49_keep_dims_0"), val = bool(true)]; tensor mu_49_cast_fp16 = reduce_mean(axes = mu_49_axes_0, keep_dims = mu_49_keep_dims_0, x = x_133_cast_fp16)[name = string("mu_49_cast_fp16")]; tensor var_3785_cast_fp16 = sub(x = x_133_cast_fp16, y = mu_49_cast_fp16)[name = string("op_3785_cast_fp16")]; fp16 var_3774_promoted_to_fp16 = const()[name = string("op_3774_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_3786_cast_fp16 = pow(x = var_3785_cast_fp16, y = var_3774_promoted_to_fp16)[name = string("op_3786_cast_fp16")]; tensor var_49_axes_0 = const()[name = string("var_49_axes_0"), val = tensor([1])]; bool var_49_keep_dims_0 = const()[name = string("var_49_keep_dims_0"), val = bool(true)]; tensor var_49_cast_fp16 = reduce_mean(axes = var_49_axes_0, keep_dims = var_49_keep_dims_0, x = var_3786_cast_fp16)[name = string("var_49_cast_fp16")]; fp16 var_3790_to_fp16 = const()[name = string("op_3790_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_3791_cast_fp16 = add(x = var_49_cast_fp16, y = var_3790_to_fp16)[name = string("op_3791_cast_fp16")]; fp32 var_3792_epsilon_0 = const()[name = string("op_3792_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_3792_cast_fp16 = rsqrt(epsilon = var_3792_epsilon_0, x = var_3791_cast_fp16)[name = string("op_3792_cast_fp16")]; tensor x_135_cast_fp16 = mul(x = var_3785_cast_fp16, y = var_3792_cast_fp16)[name = string("x_135_cast_fp16")]; tensor input_105_gamma_0_to_fp16 = const()[name = string("input_105_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(326392256)))]; tensor input_105_beta_0_to_fp16 = const()[name = string("input_105_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(326394368)))]; fp16 input_105_epsilon_0_to_fp16 = const()[name = string("input_105_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_105_cast_fp16 = batch_norm(beta = input_105_beta_0_to_fp16, epsilon = input_105_epsilon_0_to_fp16, gamma = input_105_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_135_cast_fp16)[name = string("input_105_cast_fp16")]; string var_3810_pad_type_0 = const()[name = string("op_3810_pad_type_0"), val = string("valid")]; tensor var_3810_strides_0 = const()[name = string("op_3810_strides_0"), val = tensor([1, 1])]; tensor var_3810_pad_0 = const()[name = string("op_3810_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3810_dilations_0 = const()[name = string("op_3810_dilations_0"), val = tensor([1, 1])]; int32 var_3810_groups_0 = const()[name = string("op_3810_groups_0"), val = int32(1)]; tensor var_3812_weight_0_to_fp16 = const()[name = string("op_3812_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(326396480)))]; tensor var_3812_bias_0_to_fp16 = const()[name = string("op_3812_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(328493696)))]; tensor var_3812_cast_fp16 = conv(bias = var_3812_bias_0_to_fp16, dilations = var_3810_dilations_0, groups = var_3810_groups_0, pad = var_3810_pad_0, pad_type = var_3810_pad_type_0, strides = var_3810_strides_0, weight = var_3812_weight_0_to_fp16, x = input_105_cast_fp16)[name = string("op_3812_cast_fp16")]; string var_3819_pad_type_0 = const()[name = string("op_3819_pad_type_0"), val = string("valid")]; tensor var_3819_strides_0 = const()[name = string("op_3819_strides_0"), val = tensor([1, 1])]; tensor var_3819_pad_0 = const()[name = string("op_3819_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3819_dilations_0 = const()[name = string("op_3819_dilations_0"), val = tensor([1, 1])]; int32 var_3819_groups_0 = const()[name = string("op_3819_groups_0"), val = int32(1)]; tensor layers_12_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_12_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(328495808)))]; tensor layers_12_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_12_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(330593024)))]; tensor var_3819_cast_fp16 = conv(bias = layers_12_self_attn_k_proj_bias_to_fp16, dilations = var_3819_dilations_0, groups = var_3819_groups_0, pad = var_3819_pad_0, pad_type = var_3819_pad_type_0, strides = var_3819_strides_0, weight = layers_12_self_attn_k_proj_weight_to_fp16, x = input_105_cast_fp16)[name = string("op_3819_cast_fp16")]; string var_3826_pad_type_0 = const()[name = string("op_3826_pad_type_0"), val = string("valid")]; tensor var_3826_strides_0 = const()[name = string("op_3826_strides_0"), val = tensor([1, 1])]; tensor var_3826_pad_0 = const()[name = string("op_3826_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_3826_dilations_0 = const()[name = string("op_3826_dilations_0"), val = tensor([1, 1])]; int32 var_3826_groups_0 = const()[name = string("op_3826_groups_0"), val = int32(1)]; tensor layers_12_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_12_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(330595136)))]; tensor layers_12_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_12_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(332692352)))]; tensor var_3826_cast_fp16 = conv(bias = layers_12_self_attn_v_proj_bias_to_fp16, dilations = var_3826_dilations_0, groups = var_3826_groups_0, pad = var_3826_pad_0, pad_type = var_3826_pad_type_0, strides = var_3826_strides_0, weight = layers_12_self_attn_v_proj_weight_to_fp16, x = input_105_cast_fp16)[name = string("op_3826_cast_fp16")]; tensor tile_36 = const()[name = string("tile_36"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(332694464)))]; int32 var_3827_axis_0 = const()[name = string("op_3827_axis_0"), val = int32(1)]; tensor var_3827_cast_fp16_0, tensor var_3827_cast_fp16_1, tensor var_3827_cast_fp16_2, tensor var_3827_cast_fp16_3, tensor var_3827_cast_fp16_4, tensor var_3827_cast_fp16_5, tensor var_3827_cast_fp16_6, tensor var_3827_cast_fp16_7, tensor var_3827_cast_fp16_8, tensor var_3827_cast_fp16_9, tensor var_3827_cast_fp16_10, tensor var_3827_cast_fp16_11, tensor var_3827_cast_fp16_12, tensor var_3827_cast_fp16_13, tensor var_3827_cast_fp16_14, tensor var_3827_cast_fp16_15 = split(axis = var_3827_axis_0, split_sizes = tile_36, x = var_3812_cast_fp16)[name = string("op_3827_cast_fp16")]; tensor tile_37 = const()[name = string("tile_37"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(332694592)))]; int32 var_3844_axis_0 = const()[name = string("op_3844_axis_0"), val = int32(1)]; tensor var_3844_cast_fp16_0, tensor var_3844_cast_fp16_1, tensor var_3844_cast_fp16_2, tensor var_3844_cast_fp16_3, tensor var_3844_cast_fp16_4, tensor var_3844_cast_fp16_5, tensor var_3844_cast_fp16_6, tensor var_3844_cast_fp16_7, tensor var_3844_cast_fp16_8, tensor var_3844_cast_fp16_9, tensor var_3844_cast_fp16_10, tensor var_3844_cast_fp16_11, tensor var_3844_cast_fp16_12, tensor var_3844_cast_fp16_13, tensor var_3844_cast_fp16_14, tensor var_3844_cast_fp16_15 = split(axis = var_3844_axis_0, split_sizes = tile_37, x = var_3819_cast_fp16)[name = string("op_3844_cast_fp16")]; tensor tile_38 = const()[name = string("tile_38"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(332694720)))]; int32 var_3861_axis_0 = const()[name = string("op_3861_axis_0"), val = int32(1)]; tensor var_3861_cast_fp16_0, tensor var_3861_cast_fp16_1, tensor var_3861_cast_fp16_2, tensor var_3861_cast_fp16_3, tensor var_3861_cast_fp16_4, tensor var_3861_cast_fp16_5, tensor var_3861_cast_fp16_6, tensor var_3861_cast_fp16_7, tensor var_3861_cast_fp16_8, tensor var_3861_cast_fp16_9, tensor var_3861_cast_fp16_10, tensor var_3861_cast_fp16_11, tensor var_3861_cast_fp16_12, tensor var_3861_cast_fp16_13, tensor var_3861_cast_fp16_14, tensor var_3861_cast_fp16_15 = split(axis = var_3861_axis_0, split_sizes = tile_38, x = var_3826_cast_fp16)[name = string("op_3861_cast_fp16")]; tensor transpose_384_perm_0 = const()[name = string("transpose_384_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1924 = const()[name = string("concat_1924"), val = tensor([1, 104, 64])]; tensor transpose_384_cast_fp16 = transpose(perm = transpose_384_perm_0, x = var_3827_cast_fp16_0)[name = string("transpose_3647")]; tensor reshape_576_cast_fp16 = reshape(shape = concat_1924, x = transpose_384_cast_fp16)[name = string("reshape_576_cast_fp16")]; tensor transpose_385_perm_0 = const()[name = string("transpose_385_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1925 = const()[name = string("concat_1925"), val = tensor([1, 64, 104])]; tensor transpose_385_cast_fp16 = transpose(perm = transpose_385_perm_0, x = var_3844_cast_fp16_0)[name = string("transpose_3646")]; tensor reshape_577_cast_fp16 = reshape(shape = concat_1925, x = transpose_385_cast_fp16)[name = string("reshape_577_cast_fp16")]; bool matmul_192_transpose_x_0 = const()[name = string("matmul_192_transpose_x_0"), val = bool(false)]; bool matmul_192_transpose_y_0 = const()[name = string("matmul_192_transpose_y_0"), val = bool(false)]; tensor matmul_192_cast_fp16 = matmul(transpose_x = matmul_192_transpose_x_0, transpose_y = matmul_192_transpose_y_0, x = reshape_576_cast_fp16, y = reshape_577_cast_fp16)[name = string("matmul_192_cast_fp16")]; tensor concat_1929 = const()[name = string("concat_1929"), val = tensor([1, 1, 104, 104])]; tensor reshape_578_cast_fp16 = reshape(shape = concat_1929, x = matmul_192_cast_fp16)[name = string("reshape_578_cast_fp16")]; tensor transpose_2880_perm_0 = const()[name = string("transpose_2880_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2880 = transpose(perm = transpose_2880_perm_0, x = reshape_578_cast_fp16)[name = string("transpose_3645")]; tensor w_771_cast_fp16 = add(x = transpose_2880, y = transpose_2305)[name = string("w_771_cast_fp16")]; tensor var_3883_cast_fp16 = softmax(axis = var_3771, x = w_771_cast_fp16)[name = string("op_3883_cast_fp16")]; string var_3885_equation_0 = const()[name = string("op_3885_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3885_cast_fp16 = einsum(equation = var_3885_equation_0, values = (var_3861_cast_fp16_0, var_3883_cast_fp16))[name = string("op_3885_cast_fp16")]; tensor transpose_386_perm_0 = const()[name = string("transpose_386_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1934 = const()[name = string("concat_1934"), val = tensor([1, 104, 64])]; tensor transpose_386_cast_fp16 = transpose(perm = transpose_386_perm_0, x = var_3827_cast_fp16_1)[name = string("transpose_3644")]; tensor reshape_579_cast_fp16 = reshape(shape = concat_1934, x = transpose_386_cast_fp16)[name = string("reshape_579_cast_fp16")]; tensor transpose_387_perm_0 = const()[name = string("transpose_387_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1935 = const()[name = string("concat_1935"), val = tensor([1, 64, 104])]; tensor transpose_387_cast_fp16 = transpose(perm = transpose_387_perm_0, x = var_3844_cast_fp16_1)[name = string("transpose_3643")]; tensor reshape_580_cast_fp16 = reshape(shape = concat_1935, x = transpose_387_cast_fp16)[name = string("reshape_580_cast_fp16")]; bool matmul_193_transpose_x_0 = const()[name = string("matmul_193_transpose_x_0"), val = bool(false)]; bool matmul_193_transpose_y_0 = const()[name = string("matmul_193_transpose_y_0"), val = bool(false)]; tensor matmul_193_cast_fp16 = matmul(transpose_x = matmul_193_transpose_x_0, transpose_y = matmul_193_transpose_y_0, x = reshape_579_cast_fp16, y = reshape_580_cast_fp16)[name = string("matmul_193_cast_fp16")]; tensor concat_1939 = const()[name = string("concat_1939"), val = tensor([1, 1, 104, 104])]; tensor reshape_581_cast_fp16 = reshape(shape = concat_1939, x = matmul_193_cast_fp16)[name = string("reshape_581_cast_fp16")]; tensor transpose_2881_perm_0 = const()[name = string("transpose_2881_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2881 = transpose(perm = transpose_2881_perm_0, x = reshape_581_cast_fp16)[name = string("transpose_3642")]; tensor w_775_cast_fp16 = add(x = transpose_2881, y = transpose_2305)[name = string("w_775_cast_fp16")]; tensor var_3891_cast_fp16 = softmax(axis = var_3771, x = w_775_cast_fp16)[name = string("op_3891_cast_fp16")]; string var_3893_equation_0 = const()[name = string("op_3893_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3893_cast_fp16 = einsum(equation = var_3893_equation_0, values = (var_3861_cast_fp16_1, var_3891_cast_fp16))[name = string("op_3893_cast_fp16")]; tensor transpose_388_perm_0 = const()[name = string("transpose_388_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1944 = const()[name = string("concat_1944"), val = tensor([1, 104, 64])]; tensor transpose_388_cast_fp16 = transpose(perm = transpose_388_perm_0, x = var_3827_cast_fp16_2)[name = string("transpose_3641")]; tensor reshape_582_cast_fp16 = reshape(shape = concat_1944, x = transpose_388_cast_fp16)[name = string("reshape_582_cast_fp16")]; tensor transpose_389_perm_0 = const()[name = string("transpose_389_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1945 = const()[name = string("concat_1945"), val = tensor([1, 64, 104])]; tensor transpose_389_cast_fp16 = transpose(perm = transpose_389_perm_0, x = var_3844_cast_fp16_2)[name = string("transpose_3640")]; tensor reshape_583_cast_fp16 = reshape(shape = concat_1945, x = transpose_389_cast_fp16)[name = string("reshape_583_cast_fp16")]; bool matmul_194_transpose_x_0 = const()[name = string("matmul_194_transpose_x_0"), val = bool(false)]; bool matmul_194_transpose_y_0 = const()[name = string("matmul_194_transpose_y_0"), val = bool(false)]; tensor matmul_194_cast_fp16 = matmul(transpose_x = matmul_194_transpose_x_0, transpose_y = matmul_194_transpose_y_0, x = reshape_582_cast_fp16, y = reshape_583_cast_fp16)[name = string("matmul_194_cast_fp16")]; tensor concat_1949 = const()[name = string("concat_1949"), val = tensor([1, 1, 104, 104])]; tensor reshape_584_cast_fp16 = reshape(shape = concat_1949, x = matmul_194_cast_fp16)[name = string("reshape_584_cast_fp16")]; tensor transpose_2882_perm_0 = const()[name = string("transpose_2882_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2882 = transpose(perm = transpose_2882_perm_0, x = reshape_584_cast_fp16)[name = string("transpose_3639")]; tensor w_779_cast_fp16 = add(x = transpose_2882, y = transpose_2305)[name = string("w_779_cast_fp16")]; tensor var_3899_cast_fp16 = softmax(axis = var_3771, x = w_779_cast_fp16)[name = string("op_3899_cast_fp16")]; string var_3901_equation_0 = const()[name = string("op_3901_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3901_cast_fp16 = einsum(equation = var_3901_equation_0, values = (var_3861_cast_fp16_2, var_3899_cast_fp16))[name = string("op_3901_cast_fp16")]; tensor transpose_390_perm_0 = const()[name = string("transpose_390_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1954 = const()[name = string("concat_1954"), val = tensor([1, 104, 64])]; tensor transpose_390_cast_fp16 = transpose(perm = transpose_390_perm_0, x = var_3827_cast_fp16_3)[name = string("transpose_3638")]; tensor reshape_585_cast_fp16 = reshape(shape = concat_1954, x = transpose_390_cast_fp16)[name = string("reshape_585_cast_fp16")]; tensor transpose_391_perm_0 = const()[name = string("transpose_391_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1955 = const()[name = string("concat_1955"), val = tensor([1, 64, 104])]; tensor transpose_391_cast_fp16 = transpose(perm = transpose_391_perm_0, x = var_3844_cast_fp16_3)[name = string("transpose_3637")]; tensor reshape_586_cast_fp16 = reshape(shape = concat_1955, x = transpose_391_cast_fp16)[name = string("reshape_586_cast_fp16")]; bool matmul_195_transpose_x_0 = const()[name = string("matmul_195_transpose_x_0"), val = bool(false)]; bool matmul_195_transpose_y_0 = const()[name = string("matmul_195_transpose_y_0"), val = bool(false)]; tensor matmul_195_cast_fp16 = matmul(transpose_x = matmul_195_transpose_x_0, transpose_y = matmul_195_transpose_y_0, x = reshape_585_cast_fp16, y = reshape_586_cast_fp16)[name = string("matmul_195_cast_fp16")]; tensor concat_1959 = const()[name = string("concat_1959"), val = tensor([1, 1, 104, 104])]; tensor reshape_587_cast_fp16 = reshape(shape = concat_1959, x = matmul_195_cast_fp16)[name = string("reshape_587_cast_fp16")]; tensor transpose_2883_perm_0 = const()[name = string("transpose_2883_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2883 = transpose(perm = transpose_2883_perm_0, x = reshape_587_cast_fp16)[name = string("transpose_3636")]; tensor w_783_cast_fp16 = add(x = transpose_2883, y = transpose_2305)[name = string("w_783_cast_fp16")]; tensor var_3907_cast_fp16 = softmax(axis = var_3771, x = w_783_cast_fp16)[name = string("op_3907_cast_fp16")]; string var_3909_equation_0 = const()[name = string("op_3909_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3909_cast_fp16 = einsum(equation = var_3909_equation_0, values = (var_3861_cast_fp16_3, var_3907_cast_fp16))[name = string("op_3909_cast_fp16")]; tensor transpose_392_perm_0 = const()[name = string("transpose_392_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1964 = const()[name = string("concat_1964"), val = tensor([1, 104, 64])]; tensor transpose_392_cast_fp16 = transpose(perm = transpose_392_perm_0, x = var_3827_cast_fp16_4)[name = string("transpose_3635")]; tensor reshape_588_cast_fp16 = reshape(shape = concat_1964, x = transpose_392_cast_fp16)[name = string("reshape_588_cast_fp16")]; tensor transpose_393_perm_0 = const()[name = string("transpose_393_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1965 = const()[name = string("concat_1965"), val = tensor([1, 64, 104])]; tensor transpose_393_cast_fp16 = transpose(perm = transpose_393_perm_0, x = var_3844_cast_fp16_4)[name = string("transpose_3634")]; tensor reshape_589_cast_fp16 = reshape(shape = concat_1965, x = transpose_393_cast_fp16)[name = string("reshape_589_cast_fp16")]; bool matmul_196_transpose_x_0 = const()[name = string("matmul_196_transpose_x_0"), val = bool(false)]; bool matmul_196_transpose_y_0 = const()[name = string("matmul_196_transpose_y_0"), val = bool(false)]; tensor matmul_196_cast_fp16 = matmul(transpose_x = matmul_196_transpose_x_0, transpose_y = matmul_196_transpose_y_0, x = reshape_588_cast_fp16, y = reshape_589_cast_fp16)[name = string("matmul_196_cast_fp16")]; tensor concat_1969 = const()[name = string("concat_1969"), val = tensor([1, 1, 104, 104])]; tensor reshape_590_cast_fp16 = reshape(shape = concat_1969, x = matmul_196_cast_fp16)[name = string("reshape_590_cast_fp16")]; tensor transpose_2884_perm_0 = const()[name = string("transpose_2884_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2884 = transpose(perm = transpose_2884_perm_0, x = reshape_590_cast_fp16)[name = string("transpose_3633")]; tensor w_787_cast_fp16 = add(x = transpose_2884, y = transpose_2305)[name = string("w_787_cast_fp16")]; tensor var_3915_cast_fp16 = softmax(axis = var_3771, x = w_787_cast_fp16)[name = string("op_3915_cast_fp16")]; string var_3917_equation_0 = const()[name = string("op_3917_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3917_cast_fp16 = einsum(equation = var_3917_equation_0, values = (var_3861_cast_fp16_4, var_3915_cast_fp16))[name = string("op_3917_cast_fp16")]; tensor transpose_394_perm_0 = const()[name = string("transpose_394_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1974 = const()[name = string("concat_1974"), val = tensor([1, 104, 64])]; tensor transpose_394_cast_fp16 = transpose(perm = transpose_394_perm_0, x = var_3827_cast_fp16_5)[name = string("transpose_3632")]; tensor reshape_591_cast_fp16 = reshape(shape = concat_1974, x = transpose_394_cast_fp16)[name = string("reshape_591_cast_fp16")]; tensor transpose_395_perm_0 = const()[name = string("transpose_395_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1975 = const()[name = string("concat_1975"), val = tensor([1, 64, 104])]; tensor transpose_395_cast_fp16 = transpose(perm = transpose_395_perm_0, x = var_3844_cast_fp16_5)[name = string("transpose_3631")]; tensor reshape_592_cast_fp16 = reshape(shape = concat_1975, x = transpose_395_cast_fp16)[name = string("reshape_592_cast_fp16")]; bool matmul_197_transpose_x_0 = const()[name = string("matmul_197_transpose_x_0"), val = bool(false)]; bool matmul_197_transpose_y_0 = const()[name = string("matmul_197_transpose_y_0"), val = bool(false)]; tensor matmul_197_cast_fp16 = matmul(transpose_x = matmul_197_transpose_x_0, transpose_y = matmul_197_transpose_y_0, x = reshape_591_cast_fp16, y = reshape_592_cast_fp16)[name = string("matmul_197_cast_fp16")]; tensor concat_1979 = const()[name = string("concat_1979"), val = tensor([1, 1, 104, 104])]; tensor reshape_593_cast_fp16 = reshape(shape = concat_1979, x = matmul_197_cast_fp16)[name = string("reshape_593_cast_fp16")]; tensor transpose_2885_perm_0 = const()[name = string("transpose_2885_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2885 = transpose(perm = transpose_2885_perm_0, x = reshape_593_cast_fp16)[name = string("transpose_3630")]; tensor w_791_cast_fp16 = add(x = transpose_2885, y = transpose_2305)[name = string("w_791_cast_fp16")]; tensor var_3923_cast_fp16 = softmax(axis = var_3771, x = w_791_cast_fp16)[name = string("op_3923_cast_fp16")]; string var_3925_equation_0 = const()[name = string("op_3925_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3925_cast_fp16 = einsum(equation = var_3925_equation_0, values = (var_3861_cast_fp16_5, var_3923_cast_fp16))[name = string("op_3925_cast_fp16")]; tensor transpose_396_perm_0 = const()[name = string("transpose_396_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1984 = const()[name = string("concat_1984"), val = tensor([1, 104, 64])]; tensor transpose_396_cast_fp16 = transpose(perm = transpose_396_perm_0, x = var_3827_cast_fp16_6)[name = string("transpose_3629")]; tensor reshape_594_cast_fp16 = reshape(shape = concat_1984, x = transpose_396_cast_fp16)[name = string("reshape_594_cast_fp16")]; tensor transpose_397_perm_0 = const()[name = string("transpose_397_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1985 = const()[name = string("concat_1985"), val = tensor([1, 64, 104])]; tensor transpose_397_cast_fp16 = transpose(perm = transpose_397_perm_0, x = var_3844_cast_fp16_6)[name = string("transpose_3628")]; tensor reshape_595_cast_fp16 = reshape(shape = concat_1985, x = transpose_397_cast_fp16)[name = string("reshape_595_cast_fp16")]; bool matmul_198_transpose_x_0 = const()[name = string("matmul_198_transpose_x_0"), val = bool(false)]; bool matmul_198_transpose_y_0 = const()[name = string("matmul_198_transpose_y_0"), val = bool(false)]; tensor matmul_198_cast_fp16 = matmul(transpose_x = matmul_198_transpose_x_0, transpose_y = matmul_198_transpose_y_0, x = reshape_594_cast_fp16, y = reshape_595_cast_fp16)[name = string("matmul_198_cast_fp16")]; tensor concat_1989 = const()[name = string("concat_1989"), val = tensor([1, 1, 104, 104])]; tensor reshape_596_cast_fp16 = reshape(shape = concat_1989, x = matmul_198_cast_fp16)[name = string("reshape_596_cast_fp16")]; tensor transpose_2886_perm_0 = const()[name = string("transpose_2886_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2886 = transpose(perm = transpose_2886_perm_0, x = reshape_596_cast_fp16)[name = string("transpose_3627")]; tensor w_795_cast_fp16 = add(x = transpose_2886, y = transpose_2305)[name = string("w_795_cast_fp16")]; tensor var_3931_cast_fp16 = softmax(axis = var_3771, x = w_795_cast_fp16)[name = string("op_3931_cast_fp16")]; string var_3933_equation_0 = const()[name = string("op_3933_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3933_cast_fp16 = einsum(equation = var_3933_equation_0, values = (var_3861_cast_fp16_6, var_3931_cast_fp16))[name = string("op_3933_cast_fp16")]; tensor transpose_398_perm_0 = const()[name = string("transpose_398_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_1994 = const()[name = string("concat_1994"), val = tensor([1, 104, 64])]; tensor transpose_398_cast_fp16 = transpose(perm = transpose_398_perm_0, x = var_3827_cast_fp16_7)[name = string("transpose_3626")]; tensor reshape_597_cast_fp16 = reshape(shape = concat_1994, x = transpose_398_cast_fp16)[name = string("reshape_597_cast_fp16")]; tensor transpose_399_perm_0 = const()[name = string("transpose_399_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_1995 = const()[name = string("concat_1995"), val = tensor([1, 64, 104])]; tensor transpose_399_cast_fp16 = transpose(perm = transpose_399_perm_0, x = var_3844_cast_fp16_7)[name = string("transpose_3625")]; tensor reshape_598_cast_fp16 = reshape(shape = concat_1995, x = transpose_399_cast_fp16)[name = string("reshape_598_cast_fp16")]; bool matmul_199_transpose_x_0 = const()[name = string("matmul_199_transpose_x_0"), val = bool(false)]; bool matmul_199_transpose_y_0 = const()[name = string("matmul_199_transpose_y_0"), val = bool(false)]; tensor matmul_199_cast_fp16 = matmul(transpose_x = matmul_199_transpose_x_0, transpose_y = matmul_199_transpose_y_0, x = reshape_597_cast_fp16, y = reshape_598_cast_fp16)[name = string("matmul_199_cast_fp16")]; tensor concat_1999 = const()[name = string("concat_1999"), val = tensor([1, 1, 104, 104])]; tensor reshape_599_cast_fp16 = reshape(shape = concat_1999, x = matmul_199_cast_fp16)[name = string("reshape_599_cast_fp16")]; tensor transpose_2887_perm_0 = const()[name = string("transpose_2887_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2887 = transpose(perm = transpose_2887_perm_0, x = reshape_599_cast_fp16)[name = string("transpose_3624")]; tensor w_799_cast_fp16 = add(x = transpose_2887, y = transpose_2305)[name = string("w_799_cast_fp16")]; tensor var_3939_cast_fp16 = softmax(axis = var_3771, x = w_799_cast_fp16)[name = string("op_3939_cast_fp16")]; string var_3941_equation_0 = const()[name = string("op_3941_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3941_cast_fp16 = einsum(equation = var_3941_equation_0, values = (var_3861_cast_fp16_7, var_3939_cast_fp16))[name = string("op_3941_cast_fp16")]; tensor transpose_400_perm_0 = const()[name = string("transpose_400_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2004 = const()[name = string("concat_2004"), val = tensor([1, 104, 64])]; tensor transpose_400_cast_fp16 = transpose(perm = transpose_400_perm_0, x = var_3827_cast_fp16_8)[name = string("transpose_3623")]; tensor reshape_600_cast_fp16 = reshape(shape = concat_2004, x = transpose_400_cast_fp16)[name = string("reshape_600_cast_fp16")]; tensor transpose_401_perm_0 = const()[name = string("transpose_401_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2005 = const()[name = string("concat_2005"), val = tensor([1, 64, 104])]; tensor transpose_401_cast_fp16 = transpose(perm = transpose_401_perm_0, x = var_3844_cast_fp16_8)[name = string("transpose_3622")]; tensor reshape_601_cast_fp16 = reshape(shape = concat_2005, x = transpose_401_cast_fp16)[name = string("reshape_601_cast_fp16")]; bool matmul_200_transpose_x_0 = const()[name = string("matmul_200_transpose_x_0"), val = bool(false)]; bool matmul_200_transpose_y_0 = const()[name = string("matmul_200_transpose_y_0"), val = bool(false)]; tensor matmul_200_cast_fp16 = matmul(transpose_x = matmul_200_transpose_x_0, transpose_y = matmul_200_transpose_y_0, x = reshape_600_cast_fp16, y = reshape_601_cast_fp16)[name = string("matmul_200_cast_fp16")]; tensor concat_2009 = const()[name = string("concat_2009"), val = tensor([1, 1, 104, 104])]; tensor reshape_602_cast_fp16 = reshape(shape = concat_2009, x = matmul_200_cast_fp16)[name = string("reshape_602_cast_fp16")]; tensor transpose_2888_perm_0 = const()[name = string("transpose_2888_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2888 = transpose(perm = transpose_2888_perm_0, x = reshape_602_cast_fp16)[name = string("transpose_3621")]; tensor w_803_cast_fp16 = add(x = transpose_2888, y = transpose_2305)[name = string("w_803_cast_fp16")]; tensor var_3947_cast_fp16 = softmax(axis = var_3771, x = w_803_cast_fp16)[name = string("op_3947_cast_fp16")]; string var_3949_equation_0 = const()[name = string("op_3949_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3949_cast_fp16 = einsum(equation = var_3949_equation_0, values = (var_3861_cast_fp16_8, var_3947_cast_fp16))[name = string("op_3949_cast_fp16")]; tensor transpose_402_perm_0 = const()[name = string("transpose_402_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2014 = const()[name = string("concat_2014"), val = tensor([1, 104, 64])]; tensor transpose_402_cast_fp16 = transpose(perm = transpose_402_perm_0, x = var_3827_cast_fp16_9)[name = string("transpose_3620")]; tensor reshape_603_cast_fp16 = reshape(shape = concat_2014, x = transpose_402_cast_fp16)[name = string("reshape_603_cast_fp16")]; tensor transpose_403_perm_0 = const()[name = string("transpose_403_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2015 = const()[name = string("concat_2015"), val = tensor([1, 64, 104])]; tensor transpose_403_cast_fp16 = transpose(perm = transpose_403_perm_0, x = var_3844_cast_fp16_9)[name = string("transpose_3619")]; tensor reshape_604_cast_fp16 = reshape(shape = concat_2015, x = transpose_403_cast_fp16)[name = string("reshape_604_cast_fp16")]; bool matmul_201_transpose_x_0 = const()[name = string("matmul_201_transpose_x_0"), val = bool(false)]; bool matmul_201_transpose_y_0 = const()[name = string("matmul_201_transpose_y_0"), val = bool(false)]; tensor matmul_201_cast_fp16 = matmul(transpose_x = matmul_201_transpose_x_0, transpose_y = matmul_201_transpose_y_0, x = reshape_603_cast_fp16, y = reshape_604_cast_fp16)[name = string("matmul_201_cast_fp16")]; tensor concat_2019 = const()[name = string("concat_2019"), val = tensor([1, 1, 104, 104])]; tensor reshape_605_cast_fp16 = reshape(shape = concat_2019, x = matmul_201_cast_fp16)[name = string("reshape_605_cast_fp16")]; tensor transpose_2889_perm_0 = const()[name = string("transpose_2889_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2889 = transpose(perm = transpose_2889_perm_0, x = reshape_605_cast_fp16)[name = string("transpose_3618")]; tensor w_807_cast_fp16 = add(x = transpose_2889, y = transpose_2305)[name = string("w_807_cast_fp16")]; tensor var_3955_cast_fp16 = softmax(axis = var_3771, x = w_807_cast_fp16)[name = string("op_3955_cast_fp16")]; string var_3957_equation_0 = const()[name = string("op_3957_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3957_cast_fp16 = einsum(equation = var_3957_equation_0, values = (var_3861_cast_fp16_9, var_3955_cast_fp16))[name = string("op_3957_cast_fp16")]; tensor transpose_404_perm_0 = const()[name = string("transpose_404_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2024 = const()[name = string("concat_2024"), val = tensor([1, 104, 64])]; tensor transpose_404_cast_fp16 = transpose(perm = transpose_404_perm_0, x = var_3827_cast_fp16_10)[name = string("transpose_3617")]; tensor reshape_606_cast_fp16 = reshape(shape = concat_2024, x = transpose_404_cast_fp16)[name = string("reshape_606_cast_fp16")]; tensor transpose_405_perm_0 = const()[name = string("transpose_405_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2025 = const()[name = string("concat_2025"), val = tensor([1, 64, 104])]; tensor transpose_405_cast_fp16 = transpose(perm = transpose_405_perm_0, x = var_3844_cast_fp16_10)[name = string("transpose_3616")]; tensor reshape_607_cast_fp16 = reshape(shape = concat_2025, x = transpose_405_cast_fp16)[name = string("reshape_607_cast_fp16")]; bool matmul_202_transpose_x_0 = const()[name = string("matmul_202_transpose_x_0"), val = bool(false)]; bool matmul_202_transpose_y_0 = const()[name = string("matmul_202_transpose_y_0"), val = bool(false)]; tensor matmul_202_cast_fp16 = matmul(transpose_x = matmul_202_transpose_x_0, transpose_y = matmul_202_transpose_y_0, x = reshape_606_cast_fp16, y = reshape_607_cast_fp16)[name = string("matmul_202_cast_fp16")]; tensor concat_2029 = const()[name = string("concat_2029"), val = tensor([1, 1, 104, 104])]; tensor reshape_608_cast_fp16 = reshape(shape = concat_2029, x = matmul_202_cast_fp16)[name = string("reshape_608_cast_fp16")]; tensor transpose_2890_perm_0 = const()[name = string("transpose_2890_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2890 = transpose(perm = transpose_2890_perm_0, x = reshape_608_cast_fp16)[name = string("transpose_3615")]; tensor w_811_cast_fp16 = add(x = transpose_2890, y = transpose_2305)[name = string("w_811_cast_fp16")]; tensor var_3963_cast_fp16 = softmax(axis = var_3771, x = w_811_cast_fp16)[name = string("op_3963_cast_fp16")]; string var_3965_equation_0 = const()[name = string("op_3965_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3965_cast_fp16 = einsum(equation = var_3965_equation_0, values = (var_3861_cast_fp16_10, var_3963_cast_fp16))[name = string("op_3965_cast_fp16")]; tensor transpose_406_perm_0 = const()[name = string("transpose_406_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2034 = const()[name = string("concat_2034"), val = tensor([1, 104, 64])]; tensor transpose_406_cast_fp16 = transpose(perm = transpose_406_perm_0, x = var_3827_cast_fp16_11)[name = string("transpose_3614")]; tensor reshape_609_cast_fp16 = reshape(shape = concat_2034, x = transpose_406_cast_fp16)[name = string("reshape_609_cast_fp16")]; tensor transpose_407_perm_0 = const()[name = string("transpose_407_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2035 = const()[name = string("concat_2035"), val = tensor([1, 64, 104])]; tensor transpose_407_cast_fp16 = transpose(perm = transpose_407_perm_0, x = var_3844_cast_fp16_11)[name = string("transpose_3613")]; tensor reshape_610_cast_fp16 = reshape(shape = concat_2035, x = transpose_407_cast_fp16)[name = string("reshape_610_cast_fp16")]; bool matmul_203_transpose_x_0 = const()[name = string("matmul_203_transpose_x_0"), val = bool(false)]; bool matmul_203_transpose_y_0 = const()[name = string("matmul_203_transpose_y_0"), val = bool(false)]; tensor matmul_203_cast_fp16 = matmul(transpose_x = matmul_203_transpose_x_0, transpose_y = matmul_203_transpose_y_0, x = reshape_609_cast_fp16, y = reshape_610_cast_fp16)[name = string("matmul_203_cast_fp16")]; tensor concat_2039 = const()[name = string("concat_2039"), val = tensor([1, 1, 104, 104])]; tensor reshape_611_cast_fp16 = reshape(shape = concat_2039, x = matmul_203_cast_fp16)[name = string("reshape_611_cast_fp16")]; tensor transpose_2891_perm_0 = const()[name = string("transpose_2891_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2891 = transpose(perm = transpose_2891_perm_0, x = reshape_611_cast_fp16)[name = string("transpose_3612")]; tensor w_815_cast_fp16 = add(x = transpose_2891, y = transpose_2305)[name = string("w_815_cast_fp16")]; tensor var_3971_cast_fp16 = softmax(axis = var_3771, x = w_815_cast_fp16)[name = string("op_3971_cast_fp16")]; string var_3973_equation_0 = const()[name = string("op_3973_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3973_cast_fp16 = einsum(equation = var_3973_equation_0, values = (var_3861_cast_fp16_11, var_3971_cast_fp16))[name = string("op_3973_cast_fp16")]; tensor transpose_408_perm_0 = const()[name = string("transpose_408_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2044 = const()[name = string("concat_2044"), val = tensor([1, 104, 64])]; tensor transpose_408_cast_fp16 = transpose(perm = transpose_408_perm_0, x = var_3827_cast_fp16_12)[name = string("transpose_3611")]; tensor reshape_612_cast_fp16 = reshape(shape = concat_2044, x = transpose_408_cast_fp16)[name = string("reshape_612_cast_fp16")]; tensor transpose_409_perm_0 = const()[name = string("transpose_409_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2045 = const()[name = string("concat_2045"), val = tensor([1, 64, 104])]; tensor transpose_409_cast_fp16 = transpose(perm = transpose_409_perm_0, x = var_3844_cast_fp16_12)[name = string("transpose_3610")]; tensor reshape_613_cast_fp16 = reshape(shape = concat_2045, x = transpose_409_cast_fp16)[name = string("reshape_613_cast_fp16")]; bool matmul_204_transpose_x_0 = const()[name = string("matmul_204_transpose_x_0"), val = bool(false)]; bool matmul_204_transpose_y_0 = const()[name = string("matmul_204_transpose_y_0"), val = bool(false)]; tensor matmul_204_cast_fp16 = matmul(transpose_x = matmul_204_transpose_x_0, transpose_y = matmul_204_transpose_y_0, x = reshape_612_cast_fp16, y = reshape_613_cast_fp16)[name = string("matmul_204_cast_fp16")]; tensor concat_2049 = const()[name = string("concat_2049"), val = tensor([1, 1, 104, 104])]; tensor reshape_614_cast_fp16 = reshape(shape = concat_2049, x = matmul_204_cast_fp16)[name = string("reshape_614_cast_fp16")]; tensor transpose_2892_perm_0 = const()[name = string("transpose_2892_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2892 = transpose(perm = transpose_2892_perm_0, x = reshape_614_cast_fp16)[name = string("transpose_3609")]; tensor w_819_cast_fp16 = add(x = transpose_2892, y = transpose_2305)[name = string("w_819_cast_fp16")]; tensor var_3979_cast_fp16 = softmax(axis = var_3771, x = w_819_cast_fp16)[name = string("op_3979_cast_fp16")]; string var_3981_equation_0 = const()[name = string("op_3981_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3981_cast_fp16 = einsum(equation = var_3981_equation_0, values = (var_3861_cast_fp16_12, var_3979_cast_fp16))[name = string("op_3981_cast_fp16")]; tensor transpose_410_perm_0 = const()[name = string("transpose_410_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2054 = const()[name = string("concat_2054"), val = tensor([1, 104, 64])]; tensor transpose_410_cast_fp16 = transpose(perm = transpose_410_perm_0, x = var_3827_cast_fp16_13)[name = string("transpose_3608")]; tensor reshape_615_cast_fp16 = reshape(shape = concat_2054, x = transpose_410_cast_fp16)[name = string("reshape_615_cast_fp16")]; tensor transpose_411_perm_0 = const()[name = string("transpose_411_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2055 = const()[name = string("concat_2055"), val = tensor([1, 64, 104])]; tensor transpose_411_cast_fp16 = transpose(perm = transpose_411_perm_0, x = var_3844_cast_fp16_13)[name = string("transpose_3607")]; tensor reshape_616_cast_fp16 = reshape(shape = concat_2055, x = transpose_411_cast_fp16)[name = string("reshape_616_cast_fp16")]; bool matmul_205_transpose_x_0 = const()[name = string("matmul_205_transpose_x_0"), val = bool(false)]; bool matmul_205_transpose_y_0 = const()[name = string("matmul_205_transpose_y_0"), val = bool(false)]; tensor matmul_205_cast_fp16 = matmul(transpose_x = matmul_205_transpose_x_0, transpose_y = matmul_205_transpose_y_0, x = reshape_615_cast_fp16, y = reshape_616_cast_fp16)[name = string("matmul_205_cast_fp16")]; tensor concat_2059 = const()[name = string("concat_2059"), val = tensor([1, 1, 104, 104])]; tensor reshape_617_cast_fp16 = reshape(shape = concat_2059, x = matmul_205_cast_fp16)[name = string("reshape_617_cast_fp16")]; tensor transpose_2893_perm_0 = const()[name = string("transpose_2893_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2893 = transpose(perm = transpose_2893_perm_0, x = reshape_617_cast_fp16)[name = string("transpose_3606")]; tensor w_823_cast_fp16 = add(x = transpose_2893, y = transpose_2305)[name = string("w_823_cast_fp16")]; tensor var_3987_cast_fp16 = softmax(axis = var_3771, x = w_823_cast_fp16)[name = string("op_3987_cast_fp16")]; string var_3989_equation_0 = const()[name = string("op_3989_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3989_cast_fp16 = einsum(equation = var_3989_equation_0, values = (var_3861_cast_fp16_13, var_3987_cast_fp16))[name = string("op_3989_cast_fp16")]; tensor transpose_412_perm_0 = const()[name = string("transpose_412_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2064 = const()[name = string("concat_2064"), val = tensor([1, 104, 64])]; tensor transpose_412_cast_fp16 = transpose(perm = transpose_412_perm_0, x = var_3827_cast_fp16_14)[name = string("transpose_3605")]; tensor reshape_618_cast_fp16 = reshape(shape = concat_2064, x = transpose_412_cast_fp16)[name = string("reshape_618_cast_fp16")]; tensor transpose_413_perm_0 = const()[name = string("transpose_413_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2065 = const()[name = string("concat_2065"), val = tensor([1, 64, 104])]; tensor transpose_413_cast_fp16 = transpose(perm = transpose_413_perm_0, x = var_3844_cast_fp16_14)[name = string("transpose_3604")]; tensor reshape_619_cast_fp16 = reshape(shape = concat_2065, x = transpose_413_cast_fp16)[name = string("reshape_619_cast_fp16")]; bool matmul_206_transpose_x_0 = const()[name = string("matmul_206_transpose_x_0"), val = bool(false)]; bool matmul_206_transpose_y_0 = const()[name = string("matmul_206_transpose_y_0"), val = bool(false)]; tensor matmul_206_cast_fp16 = matmul(transpose_x = matmul_206_transpose_x_0, transpose_y = matmul_206_transpose_y_0, x = reshape_618_cast_fp16, y = reshape_619_cast_fp16)[name = string("matmul_206_cast_fp16")]; tensor concat_2069 = const()[name = string("concat_2069"), val = tensor([1, 1, 104, 104])]; tensor reshape_620_cast_fp16 = reshape(shape = concat_2069, x = matmul_206_cast_fp16)[name = string("reshape_620_cast_fp16")]; tensor transpose_2894_perm_0 = const()[name = string("transpose_2894_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2894 = transpose(perm = transpose_2894_perm_0, x = reshape_620_cast_fp16)[name = string("transpose_3603")]; tensor w_827_cast_fp16 = add(x = transpose_2894, y = transpose_2305)[name = string("w_827_cast_fp16")]; tensor var_3995_cast_fp16 = softmax(axis = var_3771, x = w_827_cast_fp16)[name = string("op_3995_cast_fp16")]; string var_3997_equation_0 = const()[name = string("op_3997_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_3997_cast_fp16 = einsum(equation = var_3997_equation_0, values = (var_3861_cast_fp16_14, var_3995_cast_fp16))[name = string("op_3997_cast_fp16")]; tensor transpose_414_perm_0 = const()[name = string("transpose_414_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2074 = const()[name = string("concat_2074"), val = tensor([1, 104, 64])]; tensor transpose_414_cast_fp16 = transpose(perm = transpose_414_perm_0, x = var_3827_cast_fp16_15)[name = string("transpose_3602")]; tensor reshape_621_cast_fp16 = reshape(shape = concat_2074, x = transpose_414_cast_fp16)[name = string("reshape_621_cast_fp16")]; tensor transpose_415_perm_0 = const()[name = string("transpose_415_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2075 = const()[name = string("concat_2075"), val = tensor([1, 64, 104])]; tensor transpose_415_cast_fp16 = transpose(perm = transpose_415_perm_0, x = var_3844_cast_fp16_15)[name = string("transpose_3601")]; tensor reshape_622_cast_fp16 = reshape(shape = concat_2075, x = transpose_415_cast_fp16)[name = string("reshape_622_cast_fp16")]; bool matmul_207_transpose_x_0 = const()[name = string("matmul_207_transpose_x_0"), val = bool(false)]; bool matmul_207_transpose_y_0 = const()[name = string("matmul_207_transpose_y_0"), val = bool(false)]; tensor matmul_207_cast_fp16 = matmul(transpose_x = matmul_207_transpose_x_0, transpose_y = matmul_207_transpose_y_0, x = reshape_621_cast_fp16, y = reshape_622_cast_fp16)[name = string("matmul_207_cast_fp16")]; tensor concat_2079 = const()[name = string("concat_2079"), val = tensor([1, 1, 104, 104])]; tensor reshape_623_cast_fp16 = reshape(shape = concat_2079, x = matmul_207_cast_fp16)[name = string("reshape_623_cast_fp16")]; tensor transpose_2895_perm_0 = const()[name = string("transpose_2895_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2895 = transpose(perm = transpose_2895_perm_0, x = reshape_623_cast_fp16)[name = string("transpose_3600")]; tensor w_831_cast_fp16 = add(x = transpose_2895, y = transpose_2305)[name = string("w_831_cast_fp16")]; tensor var_4003_cast_fp16 = softmax(axis = var_3771, x = w_831_cast_fp16)[name = string("op_4003_cast_fp16")]; string var_4005_equation_0 = const()[name = string("op_4005_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4005_cast_fp16 = einsum(equation = var_4005_equation_0, values = (var_3861_cast_fp16_15, var_4003_cast_fp16))[name = string("op_4005_cast_fp16")]; bool input_107_interleave_0 = const()[name = string("input_107_interleave_0"), val = bool(false)]; tensor input_107_cast_fp16 = concat(axis = var_3771, interleave = input_107_interleave_0, values = (var_3885_cast_fp16, var_3893_cast_fp16, var_3901_cast_fp16, var_3909_cast_fp16, var_3917_cast_fp16, var_3925_cast_fp16, var_3933_cast_fp16, var_3941_cast_fp16, var_3949_cast_fp16, var_3957_cast_fp16, var_3965_cast_fp16, var_3973_cast_fp16, var_3981_cast_fp16, var_3989_cast_fp16, var_3997_cast_fp16, var_4005_cast_fp16))[name = string("input_107_cast_fp16")]; string var_4014_pad_type_0 = const()[name = string("op_4014_pad_type_0"), val = string("valid")]; tensor var_4014_strides_0 = const()[name = string("op_4014_strides_0"), val = tensor([1, 1])]; tensor var_4014_pad_0 = const()[name = string("op_4014_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4014_dilations_0 = const()[name = string("op_4014_dilations_0"), val = tensor([1, 1])]; int32 var_4014_groups_0 = const()[name = string("op_4014_groups_0"), val = int32(1)]; tensor layers_12_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_12_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(332694848)))]; tensor layers_12_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_12_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(334792064)))]; tensor var_4014_cast_fp16 = conv(bias = layers_12_self_attn_out_proj_bias_to_fp16, dilations = var_4014_dilations_0, groups = var_4014_groups_0, pad = var_4014_pad_0, pad_type = var_4014_pad_type_0, strides = var_4014_strides_0, weight = layers_12_self_attn_out_proj_weight_to_fp16, x = input_107_cast_fp16)[name = string("op_4014_cast_fp16")]; tensor x_137_cast_fp16 = add(x = x_133_cast_fp16, y = var_4014_cast_fp16)[name = string("x_137_cast_fp16")]; tensor mu_51_axes_0 = const()[name = string("mu_51_axes_0"), val = tensor([1])]; bool mu_51_keep_dims_0 = const()[name = string("mu_51_keep_dims_0"), val = bool(true)]; tensor mu_51_cast_fp16 = reduce_mean(axes = mu_51_axes_0, keep_dims = mu_51_keep_dims_0, x = x_137_cast_fp16)[name = string("mu_51_cast_fp16")]; tensor var_4020_cast_fp16 = sub(x = x_137_cast_fp16, y = mu_51_cast_fp16)[name = string("op_4020_cast_fp16")]; fp16 var_3774_promoted_1_to_fp16 = const()[name = string("op_3774_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_4021_cast_fp16 = pow(x = var_4020_cast_fp16, y = var_3774_promoted_1_to_fp16)[name = string("op_4021_cast_fp16")]; tensor var_51_axes_0 = const()[name = string("var_51_axes_0"), val = tensor([1])]; bool var_51_keep_dims_0 = const()[name = string("var_51_keep_dims_0"), val = bool(true)]; tensor var_51_cast_fp16 = reduce_mean(axes = var_51_axes_0, keep_dims = var_51_keep_dims_0, x = var_4021_cast_fp16)[name = string("var_51_cast_fp16")]; fp16 var_4025_to_fp16 = const()[name = string("op_4025_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4026_cast_fp16 = add(x = var_51_cast_fp16, y = var_4025_to_fp16)[name = string("op_4026_cast_fp16")]; fp32 var_4027_epsilon_0 = const()[name = string("op_4027_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4027_cast_fp16 = rsqrt(epsilon = var_4027_epsilon_0, x = var_4026_cast_fp16)[name = string("op_4027_cast_fp16")]; tensor x_139_cast_fp16 = mul(x = var_4020_cast_fp16, y = var_4027_cast_fp16)[name = string("x_139_cast_fp16")]; tensor input_109_gamma_0_to_fp16 = const()[name = string("input_109_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(334794176)))]; tensor input_109_beta_0_to_fp16 = const()[name = string("input_109_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(334796288)))]; fp16 input_109_epsilon_0_to_fp16 = const()[name = string("input_109_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_109_cast_fp16 = batch_norm(beta = input_109_beta_0_to_fp16, epsilon = input_109_epsilon_0_to_fp16, gamma = input_109_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_139_cast_fp16)[name = string("input_109_cast_fp16")]; string x_141_pad_type_0 = const()[name = string("x_141_pad_type_0"), val = string("valid")]; tensor x_141_strides_0 = const()[name = string("x_141_strides_0"), val = tensor([1, 1])]; tensor x_141_pad_0 = const()[name = string("x_141_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_141_dilations_0 = const()[name = string("x_141_dilations_0"), val = tensor([1, 1])]; int32 x_141_groups_0 = const()[name = string("x_141_groups_0"), val = int32(1)]; tensor layers_12_fc1_weight_to_fp16 = const()[name = string("layers_12_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(334798400)))]; tensor layers_12_fc1_bias_to_fp16 = const()[name = string("layers_12_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(343187072)))]; tensor x_141_cast_fp16 = conv(bias = layers_12_fc1_bias_to_fp16, dilations = x_141_dilations_0, groups = x_141_groups_0, pad = x_141_pad_0, pad_type = x_141_pad_type_0, strides = x_141_strides_0, weight = layers_12_fc1_weight_to_fp16, x = input_109_cast_fp16)[name = string("x_141_cast_fp16")]; fp16 var_4042_to_fp16 = const()[name = string("op_4042_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_4043_cast_fp16 = mul(x = x_141_cast_fp16, y = var_4042_to_fp16)[name = string("op_4043_cast_fp16")]; tensor var_4044_cast_fp16 = mul(x = var_4043_cast_fp16, y = x_141_cast_fp16)[name = string("op_4044_cast_fp16")]; tensor var_4045_cast_fp16 = mul(x = var_4044_cast_fp16, y = x_141_cast_fp16)[name = string("op_4045_cast_fp16")]; tensor var_4046_cast_fp16 = add(x = x_141_cast_fp16, y = var_4045_cast_fp16)[name = string("op_4046_cast_fp16")]; fp16 var_4047_to_fp16 = const()[name = string("op_4047_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_31_cast_fp16 = mul(x = var_4046_cast_fp16, y = var_4047_to_fp16)[name = string("u_31_cast_fp16")]; fp16 var_4049_to_fp16 = const()[name = string("op_4049_to_fp16"), val = fp16(0x1p-1)]; tensor var_4050_cast_fp16 = mul(x = x_141_cast_fp16, y = var_4049_to_fp16)[name = string("op_4050_cast_fp16")]; tensor var_4051_cast_fp16 = tanh(x = u_31_cast_fp16)[name = string("op_4051_cast_fp16")]; fp16 var_4052_to_fp16 = const()[name = string("op_4052_to_fp16"), val = fp16(0x1p+0)]; tensor var_4053_cast_fp16 = add(x = var_4051_cast_fp16, y = var_4052_to_fp16)[name = string("op_4053_cast_fp16")]; tensor input_111_cast_fp16 = mul(x = var_4050_cast_fp16, y = var_4053_cast_fp16)[name = string("input_111_cast_fp16")]; string h_25_pad_type_0 = const()[name = string("h_25_pad_type_0"), val = string("valid")]; tensor h_25_strides_0 = const()[name = string("h_25_strides_0"), val = tensor([1, 1])]; tensor h_25_pad_0 = const()[name = string("h_25_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_25_dilations_0 = const()[name = string("h_25_dilations_0"), val = tensor([1, 1])]; int32 h_25_groups_0 = const()[name = string("h_25_groups_0"), val = int32(1)]; tensor layers_12_fc2_weight_to_fp16 = const()[name = string("layers_12_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(343195328)))]; tensor layers_12_fc2_bias_to_fp16 = const()[name = string("layers_12_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(351584000)))]; tensor h_25_cast_fp16 = conv(bias = layers_12_fc2_bias_to_fp16, dilations = h_25_dilations_0, groups = h_25_groups_0, pad = h_25_pad_0, pad_type = h_25_pad_type_0, strides = h_25_strides_0, weight = layers_12_fc2_weight_to_fp16, x = input_111_cast_fp16)[name = string("h_25_cast_fp16")]; tensor x_143_cast_fp16 = add(x = x_137_cast_fp16, y = h_25_cast_fp16)[name = string("x_143_cast_fp16")]; int32 var_4069 = const()[name = string("op_4069"), val = int32(1)]; tensor mu_53_axes_0 = const()[name = string("mu_53_axes_0"), val = tensor([1])]; bool mu_53_keep_dims_0 = const()[name = string("mu_53_keep_dims_0"), val = bool(true)]; tensor mu_53_cast_fp16 = reduce_mean(axes = mu_53_axes_0, keep_dims = mu_53_keep_dims_0, x = x_143_cast_fp16)[name = string("mu_53_cast_fp16")]; tensor var_4083_cast_fp16 = sub(x = x_143_cast_fp16, y = mu_53_cast_fp16)[name = string("op_4083_cast_fp16")]; fp16 var_4072_promoted_to_fp16 = const()[name = string("op_4072_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_4084_cast_fp16 = pow(x = var_4083_cast_fp16, y = var_4072_promoted_to_fp16)[name = string("op_4084_cast_fp16")]; tensor var_53_axes_0 = const()[name = string("var_53_axes_0"), val = tensor([1])]; bool var_53_keep_dims_0 = const()[name = string("var_53_keep_dims_0"), val = bool(true)]; tensor var_53_cast_fp16 = reduce_mean(axes = var_53_axes_0, keep_dims = var_53_keep_dims_0, x = var_4084_cast_fp16)[name = string("var_53_cast_fp16")]; fp16 var_4088_to_fp16 = const()[name = string("op_4088_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4089_cast_fp16 = add(x = var_53_cast_fp16, y = var_4088_to_fp16)[name = string("op_4089_cast_fp16")]; fp32 var_4090_epsilon_0 = const()[name = string("op_4090_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4090_cast_fp16 = rsqrt(epsilon = var_4090_epsilon_0, x = var_4089_cast_fp16)[name = string("op_4090_cast_fp16")]; tensor x_145_cast_fp16 = mul(x = var_4083_cast_fp16, y = var_4090_cast_fp16)[name = string("x_145_cast_fp16")]; tensor input_113_gamma_0_to_fp16 = const()[name = string("input_113_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(351586112)))]; tensor input_113_beta_0_to_fp16 = const()[name = string("input_113_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(351588224)))]; fp16 input_113_epsilon_0_to_fp16 = const()[name = string("input_113_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_113_cast_fp16 = batch_norm(beta = input_113_beta_0_to_fp16, epsilon = input_113_epsilon_0_to_fp16, gamma = input_113_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_145_cast_fp16)[name = string("input_113_cast_fp16")]; string var_4108_pad_type_0 = const()[name = string("op_4108_pad_type_0"), val = string("valid")]; tensor var_4108_strides_0 = const()[name = string("op_4108_strides_0"), val = tensor([1, 1])]; tensor var_4108_pad_0 = const()[name = string("op_4108_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4108_dilations_0 = const()[name = string("op_4108_dilations_0"), val = tensor([1, 1])]; int32 var_4108_groups_0 = const()[name = string("op_4108_groups_0"), val = int32(1)]; tensor var_4110_weight_0_to_fp16 = const()[name = string("op_4110_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(351590336)))]; tensor var_4110_bias_0_to_fp16 = const()[name = string("op_4110_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(353687552)))]; tensor var_4110_cast_fp16 = conv(bias = var_4110_bias_0_to_fp16, dilations = var_4108_dilations_0, groups = var_4108_groups_0, pad = var_4108_pad_0, pad_type = var_4108_pad_type_0, strides = var_4108_strides_0, weight = var_4110_weight_0_to_fp16, x = input_113_cast_fp16)[name = string("op_4110_cast_fp16")]; string var_4117_pad_type_0 = const()[name = string("op_4117_pad_type_0"), val = string("valid")]; tensor var_4117_strides_0 = const()[name = string("op_4117_strides_0"), val = tensor([1, 1])]; tensor var_4117_pad_0 = const()[name = string("op_4117_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4117_dilations_0 = const()[name = string("op_4117_dilations_0"), val = tensor([1, 1])]; int32 var_4117_groups_0 = const()[name = string("op_4117_groups_0"), val = int32(1)]; tensor layers_13_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_13_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(353689664)))]; tensor layers_13_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_13_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(355786880)))]; tensor var_4117_cast_fp16 = conv(bias = layers_13_self_attn_k_proj_bias_to_fp16, dilations = var_4117_dilations_0, groups = var_4117_groups_0, pad = var_4117_pad_0, pad_type = var_4117_pad_type_0, strides = var_4117_strides_0, weight = layers_13_self_attn_k_proj_weight_to_fp16, x = input_113_cast_fp16)[name = string("op_4117_cast_fp16")]; string var_4124_pad_type_0 = const()[name = string("op_4124_pad_type_0"), val = string("valid")]; tensor var_4124_strides_0 = const()[name = string("op_4124_strides_0"), val = tensor([1, 1])]; tensor var_4124_pad_0 = const()[name = string("op_4124_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4124_dilations_0 = const()[name = string("op_4124_dilations_0"), val = tensor([1, 1])]; int32 var_4124_groups_0 = const()[name = string("op_4124_groups_0"), val = int32(1)]; tensor layers_13_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_13_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(355788992)))]; tensor layers_13_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_13_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(357886208)))]; tensor var_4124_cast_fp16 = conv(bias = layers_13_self_attn_v_proj_bias_to_fp16, dilations = var_4124_dilations_0, groups = var_4124_groups_0, pad = var_4124_pad_0, pad_type = var_4124_pad_type_0, strides = var_4124_strides_0, weight = layers_13_self_attn_v_proj_weight_to_fp16, x = input_113_cast_fp16)[name = string("op_4124_cast_fp16")]; tensor tile_39 = const()[name = string("tile_39"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(357888320)))]; int32 var_4125_axis_0 = const()[name = string("op_4125_axis_0"), val = int32(1)]; tensor var_4125_cast_fp16_0, tensor var_4125_cast_fp16_1, tensor var_4125_cast_fp16_2, tensor var_4125_cast_fp16_3, tensor var_4125_cast_fp16_4, tensor var_4125_cast_fp16_5, tensor var_4125_cast_fp16_6, tensor var_4125_cast_fp16_7, tensor var_4125_cast_fp16_8, tensor var_4125_cast_fp16_9, tensor var_4125_cast_fp16_10, tensor var_4125_cast_fp16_11, tensor var_4125_cast_fp16_12, tensor var_4125_cast_fp16_13, tensor var_4125_cast_fp16_14, tensor var_4125_cast_fp16_15 = split(axis = var_4125_axis_0, split_sizes = tile_39, x = var_4110_cast_fp16)[name = string("op_4125_cast_fp16")]; tensor tile_40 = const()[name = string("tile_40"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(357888448)))]; int32 var_4142_axis_0 = const()[name = string("op_4142_axis_0"), val = int32(1)]; tensor var_4142_cast_fp16_0, tensor var_4142_cast_fp16_1, tensor var_4142_cast_fp16_2, tensor var_4142_cast_fp16_3, tensor var_4142_cast_fp16_4, tensor var_4142_cast_fp16_5, tensor var_4142_cast_fp16_6, tensor var_4142_cast_fp16_7, tensor var_4142_cast_fp16_8, tensor var_4142_cast_fp16_9, tensor var_4142_cast_fp16_10, tensor var_4142_cast_fp16_11, tensor var_4142_cast_fp16_12, tensor var_4142_cast_fp16_13, tensor var_4142_cast_fp16_14, tensor var_4142_cast_fp16_15 = split(axis = var_4142_axis_0, split_sizes = tile_40, x = var_4117_cast_fp16)[name = string("op_4142_cast_fp16")]; tensor tile_41 = const()[name = string("tile_41"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(357888576)))]; int32 var_4159_axis_0 = const()[name = string("op_4159_axis_0"), val = int32(1)]; tensor var_4159_cast_fp16_0, tensor var_4159_cast_fp16_1, tensor var_4159_cast_fp16_2, tensor var_4159_cast_fp16_3, tensor var_4159_cast_fp16_4, tensor var_4159_cast_fp16_5, tensor var_4159_cast_fp16_6, tensor var_4159_cast_fp16_7, tensor var_4159_cast_fp16_8, tensor var_4159_cast_fp16_9, tensor var_4159_cast_fp16_10, tensor var_4159_cast_fp16_11, tensor var_4159_cast_fp16_12, tensor var_4159_cast_fp16_13, tensor var_4159_cast_fp16_14, tensor var_4159_cast_fp16_15 = split(axis = var_4159_axis_0, split_sizes = tile_41, x = var_4124_cast_fp16)[name = string("op_4159_cast_fp16")]; tensor transpose_416_perm_0 = const()[name = string("transpose_416_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2084 = const()[name = string("concat_2084"), val = tensor([1, 104, 64])]; tensor transpose_416_cast_fp16 = transpose(perm = transpose_416_perm_0, x = var_4125_cast_fp16_0)[name = string("transpose_3599")]; tensor reshape_624_cast_fp16 = reshape(shape = concat_2084, x = transpose_416_cast_fp16)[name = string("reshape_624_cast_fp16")]; tensor transpose_417_perm_0 = const()[name = string("transpose_417_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2085 = const()[name = string("concat_2085"), val = tensor([1, 64, 104])]; tensor transpose_417_cast_fp16 = transpose(perm = transpose_417_perm_0, x = var_4142_cast_fp16_0)[name = string("transpose_3598")]; tensor reshape_625_cast_fp16 = reshape(shape = concat_2085, x = transpose_417_cast_fp16)[name = string("reshape_625_cast_fp16")]; bool matmul_208_transpose_x_0 = const()[name = string("matmul_208_transpose_x_0"), val = bool(false)]; bool matmul_208_transpose_y_0 = const()[name = string("matmul_208_transpose_y_0"), val = bool(false)]; tensor matmul_208_cast_fp16 = matmul(transpose_x = matmul_208_transpose_x_0, transpose_y = matmul_208_transpose_y_0, x = reshape_624_cast_fp16, y = reshape_625_cast_fp16)[name = string("matmul_208_cast_fp16")]; tensor concat_2089 = const()[name = string("concat_2089"), val = tensor([1, 1, 104, 104])]; tensor reshape_626_cast_fp16 = reshape(shape = concat_2089, x = matmul_208_cast_fp16)[name = string("reshape_626_cast_fp16")]; tensor transpose_2896_perm_0 = const()[name = string("transpose_2896_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2896 = transpose(perm = transpose_2896_perm_0, x = reshape_626_cast_fp16)[name = string("transpose_3597")]; tensor w_835_cast_fp16 = add(x = transpose_2896, y = transpose_2305)[name = string("w_835_cast_fp16")]; tensor var_4181_cast_fp16 = softmax(axis = var_4069, x = w_835_cast_fp16)[name = string("op_4181_cast_fp16")]; string var_4183_equation_0 = const()[name = string("op_4183_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4183_cast_fp16 = einsum(equation = var_4183_equation_0, values = (var_4159_cast_fp16_0, var_4181_cast_fp16))[name = string("op_4183_cast_fp16")]; tensor transpose_418_perm_0 = const()[name = string("transpose_418_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2094 = const()[name = string("concat_2094"), val = tensor([1, 104, 64])]; tensor transpose_418_cast_fp16 = transpose(perm = transpose_418_perm_0, x = var_4125_cast_fp16_1)[name = string("transpose_3596")]; tensor reshape_627_cast_fp16 = reshape(shape = concat_2094, x = transpose_418_cast_fp16)[name = string("reshape_627_cast_fp16")]; tensor transpose_419_perm_0 = const()[name = string("transpose_419_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2095 = const()[name = string("concat_2095"), val = tensor([1, 64, 104])]; tensor transpose_419_cast_fp16 = transpose(perm = transpose_419_perm_0, x = var_4142_cast_fp16_1)[name = string("transpose_3595")]; tensor reshape_628_cast_fp16 = reshape(shape = concat_2095, x = transpose_419_cast_fp16)[name = string("reshape_628_cast_fp16")]; bool matmul_209_transpose_x_0 = const()[name = string("matmul_209_transpose_x_0"), val = bool(false)]; bool matmul_209_transpose_y_0 = const()[name = string("matmul_209_transpose_y_0"), val = bool(false)]; tensor matmul_209_cast_fp16 = matmul(transpose_x = matmul_209_transpose_x_0, transpose_y = matmul_209_transpose_y_0, x = reshape_627_cast_fp16, y = reshape_628_cast_fp16)[name = string("matmul_209_cast_fp16")]; tensor concat_2099 = const()[name = string("concat_2099"), val = tensor([1, 1, 104, 104])]; tensor reshape_629_cast_fp16 = reshape(shape = concat_2099, x = matmul_209_cast_fp16)[name = string("reshape_629_cast_fp16")]; tensor transpose_2897_perm_0 = const()[name = string("transpose_2897_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2897 = transpose(perm = transpose_2897_perm_0, x = reshape_629_cast_fp16)[name = string("transpose_3594")]; tensor w_839_cast_fp16 = add(x = transpose_2897, y = transpose_2305)[name = string("w_839_cast_fp16")]; tensor var_4189_cast_fp16 = softmax(axis = var_4069, x = w_839_cast_fp16)[name = string("op_4189_cast_fp16")]; string var_4191_equation_0 = const()[name = string("op_4191_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4191_cast_fp16 = einsum(equation = var_4191_equation_0, values = (var_4159_cast_fp16_1, var_4189_cast_fp16))[name = string("op_4191_cast_fp16")]; tensor transpose_420_perm_0 = const()[name = string("transpose_420_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2104 = const()[name = string("concat_2104"), val = tensor([1, 104, 64])]; tensor transpose_420_cast_fp16 = transpose(perm = transpose_420_perm_0, x = var_4125_cast_fp16_2)[name = string("transpose_3593")]; tensor reshape_630_cast_fp16 = reshape(shape = concat_2104, x = transpose_420_cast_fp16)[name = string("reshape_630_cast_fp16")]; tensor transpose_421_perm_0 = const()[name = string("transpose_421_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2105 = const()[name = string("concat_2105"), val = tensor([1, 64, 104])]; tensor transpose_421_cast_fp16 = transpose(perm = transpose_421_perm_0, x = var_4142_cast_fp16_2)[name = string("transpose_3592")]; tensor reshape_631_cast_fp16 = reshape(shape = concat_2105, x = transpose_421_cast_fp16)[name = string("reshape_631_cast_fp16")]; bool matmul_210_transpose_x_0 = const()[name = string("matmul_210_transpose_x_0"), val = bool(false)]; bool matmul_210_transpose_y_0 = const()[name = string("matmul_210_transpose_y_0"), val = bool(false)]; tensor matmul_210_cast_fp16 = matmul(transpose_x = matmul_210_transpose_x_0, transpose_y = matmul_210_transpose_y_0, x = reshape_630_cast_fp16, y = reshape_631_cast_fp16)[name = string("matmul_210_cast_fp16")]; tensor concat_2109 = const()[name = string("concat_2109"), val = tensor([1, 1, 104, 104])]; tensor reshape_632_cast_fp16 = reshape(shape = concat_2109, x = matmul_210_cast_fp16)[name = string("reshape_632_cast_fp16")]; tensor transpose_2898_perm_0 = const()[name = string("transpose_2898_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2898 = transpose(perm = transpose_2898_perm_0, x = reshape_632_cast_fp16)[name = string("transpose_3591")]; tensor w_843_cast_fp16 = add(x = transpose_2898, y = transpose_2305)[name = string("w_843_cast_fp16")]; tensor var_4197_cast_fp16 = softmax(axis = var_4069, x = w_843_cast_fp16)[name = string("op_4197_cast_fp16")]; string var_4199_equation_0 = const()[name = string("op_4199_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4199_cast_fp16 = einsum(equation = var_4199_equation_0, values = (var_4159_cast_fp16_2, var_4197_cast_fp16))[name = string("op_4199_cast_fp16")]; tensor transpose_422_perm_0 = const()[name = string("transpose_422_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2114 = const()[name = string("concat_2114"), val = tensor([1, 104, 64])]; tensor transpose_422_cast_fp16 = transpose(perm = transpose_422_perm_0, x = var_4125_cast_fp16_3)[name = string("transpose_3590")]; tensor reshape_633_cast_fp16 = reshape(shape = concat_2114, x = transpose_422_cast_fp16)[name = string("reshape_633_cast_fp16")]; tensor transpose_423_perm_0 = const()[name = string("transpose_423_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2115 = const()[name = string("concat_2115"), val = tensor([1, 64, 104])]; tensor transpose_423_cast_fp16 = transpose(perm = transpose_423_perm_0, x = var_4142_cast_fp16_3)[name = string("transpose_3589")]; tensor reshape_634_cast_fp16 = reshape(shape = concat_2115, x = transpose_423_cast_fp16)[name = string("reshape_634_cast_fp16")]; bool matmul_211_transpose_x_0 = const()[name = string("matmul_211_transpose_x_0"), val = bool(false)]; bool matmul_211_transpose_y_0 = const()[name = string("matmul_211_transpose_y_0"), val = bool(false)]; tensor matmul_211_cast_fp16 = matmul(transpose_x = matmul_211_transpose_x_0, transpose_y = matmul_211_transpose_y_0, x = reshape_633_cast_fp16, y = reshape_634_cast_fp16)[name = string("matmul_211_cast_fp16")]; tensor concat_2119 = const()[name = string("concat_2119"), val = tensor([1, 1, 104, 104])]; tensor reshape_635_cast_fp16 = reshape(shape = concat_2119, x = matmul_211_cast_fp16)[name = string("reshape_635_cast_fp16")]; tensor transpose_2899_perm_0 = const()[name = string("transpose_2899_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2899 = transpose(perm = transpose_2899_perm_0, x = reshape_635_cast_fp16)[name = string("transpose_3588")]; tensor w_847_cast_fp16 = add(x = transpose_2899, y = transpose_2305)[name = string("w_847_cast_fp16")]; tensor var_4205_cast_fp16 = softmax(axis = var_4069, x = w_847_cast_fp16)[name = string("op_4205_cast_fp16")]; string var_4207_equation_0 = const()[name = string("op_4207_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4207_cast_fp16 = einsum(equation = var_4207_equation_0, values = (var_4159_cast_fp16_3, var_4205_cast_fp16))[name = string("op_4207_cast_fp16")]; tensor transpose_424_perm_0 = const()[name = string("transpose_424_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2124 = const()[name = string("concat_2124"), val = tensor([1, 104, 64])]; tensor transpose_424_cast_fp16 = transpose(perm = transpose_424_perm_0, x = var_4125_cast_fp16_4)[name = string("transpose_3587")]; tensor reshape_636_cast_fp16 = reshape(shape = concat_2124, x = transpose_424_cast_fp16)[name = string("reshape_636_cast_fp16")]; tensor transpose_425_perm_0 = const()[name = string("transpose_425_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2125 = const()[name = string("concat_2125"), val = tensor([1, 64, 104])]; tensor transpose_425_cast_fp16 = transpose(perm = transpose_425_perm_0, x = var_4142_cast_fp16_4)[name = string("transpose_3586")]; tensor reshape_637_cast_fp16 = reshape(shape = concat_2125, x = transpose_425_cast_fp16)[name = string("reshape_637_cast_fp16")]; bool matmul_212_transpose_x_0 = const()[name = string("matmul_212_transpose_x_0"), val = bool(false)]; bool matmul_212_transpose_y_0 = const()[name = string("matmul_212_transpose_y_0"), val = bool(false)]; tensor matmul_212_cast_fp16 = matmul(transpose_x = matmul_212_transpose_x_0, transpose_y = matmul_212_transpose_y_0, x = reshape_636_cast_fp16, y = reshape_637_cast_fp16)[name = string("matmul_212_cast_fp16")]; tensor concat_2129 = const()[name = string("concat_2129"), val = tensor([1, 1, 104, 104])]; tensor reshape_638_cast_fp16 = reshape(shape = concat_2129, x = matmul_212_cast_fp16)[name = string("reshape_638_cast_fp16")]; tensor transpose_2900_perm_0 = const()[name = string("transpose_2900_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2900 = transpose(perm = transpose_2900_perm_0, x = reshape_638_cast_fp16)[name = string("transpose_3585")]; tensor w_851_cast_fp16 = add(x = transpose_2900, y = transpose_2305)[name = string("w_851_cast_fp16")]; tensor var_4213_cast_fp16 = softmax(axis = var_4069, x = w_851_cast_fp16)[name = string("op_4213_cast_fp16")]; string var_4215_equation_0 = const()[name = string("op_4215_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4215_cast_fp16 = einsum(equation = var_4215_equation_0, values = (var_4159_cast_fp16_4, var_4213_cast_fp16))[name = string("op_4215_cast_fp16")]; tensor transpose_426_perm_0 = const()[name = string("transpose_426_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2134 = const()[name = string("concat_2134"), val = tensor([1, 104, 64])]; tensor transpose_426_cast_fp16 = transpose(perm = transpose_426_perm_0, x = var_4125_cast_fp16_5)[name = string("transpose_3584")]; tensor reshape_639_cast_fp16 = reshape(shape = concat_2134, x = transpose_426_cast_fp16)[name = string("reshape_639_cast_fp16")]; tensor transpose_427_perm_0 = const()[name = string("transpose_427_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2135 = const()[name = string("concat_2135"), val = tensor([1, 64, 104])]; tensor transpose_427_cast_fp16 = transpose(perm = transpose_427_perm_0, x = var_4142_cast_fp16_5)[name = string("transpose_3583")]; tensor reshape_640_cast_fp16 = reshape(shape = concat_2135, x = transpose_427_cast_fp16)[name = string("reshape_640_cast_fp16")]; bool matmul_213_transpose_x_0 = const()[name = string("matmul_213_transpose_x_0"), val = bool(false)]; bool matmul_213_transpose_y_0 = const()[name = string("matmul_213_transpose_y_0"), val = bool(false)]; tensor matmul_213_cast_fp16 = matmul(transpose_x = matmul_213_transpose_x_0, transpose_y = matmul_213_transpose_y_0, x = reshape_639_cast_fp16, y = reshape_640_cast_fp16)[name = string("matmul_213_cast_fp16")]; tensor concat_2139 = const()[name = string("concat_2139"), val = tensor([1, 1, 104, 104])]; tensor reshape_641_cast_fp16 = reshape(shape = concat_2139, x = matmul_213_cast_fp16)[name = string("reshape_641_cast_fp16")]; tensor transpose_2901_perm_0 = const()[name = string("transpose_2901_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2901 = transpose(perm = transpose_2901_perm_0, x = reshape_641_cast_fp16)[name = string("transpose_3582")]; tensor w_855_cast_fp16 = add(x = transpose_2901, y = transpose_2305)[name = string("w_855_cast_fp16")]; tensor var_4221_cast_fp16 = softmax(axis = var_4069, x = w_855_cast_fp16)[name = string("op_4221_cast_fp16")]; string var_4223_equation_0 = const()[name = string("op_4223_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4223_cast_fp16 = einsum(equation = var_4223_equation_0, values = (var_4159_cast_fp16_5, var_4221_cast_fp16))[name = string("op_4223_cast_fp16")]; tensor transpose_428_perm_0 = const()[name = string("transpose_428_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2144 = const()[name = string("concat_2144"), val = tensor([1, 104, 64])]; tensor transpose_428_cast_fp16 = transpose(perm = transpose_428_perm_0, x = var_4125_cast_fp16_6)[name = string("transpose_3581")]; tensor reshape_642_cast_fp16 = reshape(shape = concat_2144, x = transpose_428_cast_fp16)[name = string("reshape_642_cast_fp16")]; tensor transpose_429_perm_0 = const()[name = string("transpose_429_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2145 = const()[name = string("concat_2145"), val = tensor([1, 64, 104])]; tensor transpose_429_cast_fp16 = transpose(perm = transpose_429_perm_0, x = var_4142_cast_fp16_6)[name = string("transpose_3580")]; tensor reshape_643_cast_fp16 = reshape(shape = concat_2145, x = transpose_429_cast_fp16)[name = string("reshape_643_cast_fp16")]; bool matmul_214_transpose_x_0 = const()[name = string("matmul_214_transpose_x_0"), val = bool(false)]; bool matmul_214_transpose_y_0 = const()[name = string("matmul_214_transpose_y_0"), val = bool(false)]; tensor matmul_214_cast_fp16 = matmul(transpose_x = matmul_214_transpose_x_0, transpose_y = matmul_214_transpose_y_0, x = reshape_642_cast_fp16, y = reshape_643_cast_fp16)[name = string("matmul_214_cast_fp16")]; tensor concat_2149 = const()[name = string("concat_2149"), val = tensor([1, 1, 104, 104])]; tensor reshape_644_cast_fp16 = reshape(shape = concat_2149, x = matmul_214_cast_fp16)[name = string("reshape_644_cast_fp16")]; tensor transpose_2902_perm_0 = const()[name = string("transpose_2902_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2902 = transpose(perm = transpose_2902_perm_0, x = reshape_644_cast_fp16)[name = string("transpose_3579")]; tensor w_859_cast_fp16 = add(x = transpose_2902, y = transpose_2305)[name = string("w_859_cast_fp16")]; tensor var_4229_cast_fp16 = softmax(axis = var_4069, x = w_859_cast_fp16)[name = string("op_4229_cast_fp16")]; string var_4231_equation_0 = const()[name = string("op_4231_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4231_cast_fp16 = einsum(equation = var_4231_equation_0, values = (var_4159_cast_fp16_6, var_4229_cast_fp16))[name = string("op_4231_cast_fp16")]; tensor transpose_430_perm_0 = const()[name = string("transpose_430_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2154 = const()[name = string("concat_2154"), val = tensor([1, 104, 64])]; tensor transpose_430_cast_fp16 = transpose(perm = transpose_430_perm_0, x = var_4125_cast_fp16_7)[name = string("transpose_3578")]; tensor reshape_645_cast_fp16 = reshape(shape = concat_2154, x = transpose_430_cast_fp16)[name = string("reshape_645_cast_fp16")]; tensor transpose_431_perm_0 = const()[name = string("transpose_431_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2155 = const()[name = string("concat_2155"), val = tensor([1, 64, 104])]; tensor transpose_431_cast_fp16 = transpose(perm = transpose_431_perm_0, x = var_4142_cast_fp16_7)[name = string("transpose_3577")]; tensor reshape_646_cast_fp16 = reshape(shape = concat_2155, x = transpose_431_cast_fp16)[name = string("reshape_646_cast_fp16")]; bool matmul_215_transpose_x_0 = const()[name = string("matmul_215_transpose_x_0"), val = bool(false)]; bool matmul_215_transpose_y_0 = const()[name = string("matmul_215_transpose_y_0"), val = bool(false)]; tensor matmul_215_cast_fp16 = matmul(transpose_x = matmul_215_transpose_x_0, transpose_y = matmul_215_transpose_y_0, x = reshape_645_cast_fp16, y = reshape_646_cast_fp16)[name = string("matmul_215_cast_fp16")]; tensor concat_2159 = const()[name = string("concat_2159"), val = tensor([1, 1, 104, 104])]; tensor reshape_647_cast_fp16 = reshape(shape = concat_2159, x = matmul_215_cast_fp16)[name = string("reshape_647_cast_fp16")]; tensor transpose_2903_perm_0 = const()[name = string("transpose_2903_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2903 = transpose(perm = transpose_2903_perm_0, x = reshape_647_cast_fp16)[name = string("transpose_3576")]; tensor w_863_cast_fp16 = add(x = transpose_2903, y = transpose_2305)[name = string("w_863_cast_fp16")]; tensor var_4237_cast_fp16 = softmax(axis = var_4069, x = w_863_cast_fp16)[name = string("op_4237_cast_fp16")]; string var_4239_equation_0 = const()[name = string("op_4239_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4239_cast_fp16 = einsum(equation = var_4239_equation_0, values = (var_4159_cast_fp16_7, var_4237_cast_fp16))[name = string("op_4239_cast_fp16")]; tensor transpose_432_perm_0 = const()[name = string("transpose_432_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2164 = const()[name = string("concat_2164"), val = tensor([1, 104, 64])]; tensor transpose_432_cast_fp16 = transpose(perm = transpose_432_perm_0, x = var_4125_cast_fp16_8)[name = string("transpose_3575")]; tensor reshape_648_cast_fp16 = reshape(shape = concat_2164, x = transpose_432_cast_fp16)[name = string("reshape_648_cast_fp16")]; tensor transpose_433_perm_0 = const()[name = string("transpose_433_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2165 = const()[name = string("concat_2165"), val = tensor([1, 64, 104])]; tensor transpose_433_cast_fp16 = transpose(perm = transpose_433_perm_0, x = var_4142_cast_fp16_8)[name = string("transpose_3574")]; tensor reshape_649_cast_fp16 = reshape(shape = concat_2165, x = transpose_433_cast_fp16)[name = string("reshape_649_cast_fp16")]; bool matmul_216_transpose_x_0 = const()[name = string("matmul_216_transpose_x_0"), val = bool(false)]; bool matmul_216_transpose_y_0 = const()[name = string("matmul_216_transpose_y_0"), val = bool(false)]; tensor matmul_216_cast_fp16 = matmul(transpose_x = matmul_216_transpose_x_0, transpose_y = matmul_216_transpose_y_0, x = reshape_648_cast_fp16, y = reshape_649_cast_fp16)[name = string("matmul_216_cast_fp16")]; tensor concat_2169 = const()[name = string("concat_2169"), val = tensor([1, 1, 104, 104])]; tensor reshape_650_cast_fp16 = reshape(shape = concat_2169, x = matmul_216_cast_fp16)[name = string("reshape_650_cast_fp16")]; tensor transpose_2904_perm_0 = const()[name = string("transpose_2904_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2904 = transpose(perm = transpose_2904_perm_0, x = reshape_650_cast_fp16)[name = string("transpose_3573")]; tensor w_867_cast_fp16 = add(x = transpose_2904, y = transpose_2305)[name = string("w_867_cast_fp16")]; tensor var_4245_cast_fp16 = softmax(axis = var_4069, x = w_867_cast_fp16)[name = string("op_4245_cast_fp16")]; string var_4247_equation_0 = const()[name = string("op_4247_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4247_cast_fp16 = einsum(equation = var_4247_equation_0, values = (var_4159_cast_fp16_8, var_4245_cast_fp16))[name = string("op_4247_cast_fp16")]; tensor transpose_434_perm_0 = const()[name = string("transpose_434_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2174 = const()[name = string("concat_2174"), val = tensor([1, 104, 64])]; tensor transpose_434_cast_fp16 = transpose(perm = transpose_434_perm_0, x = var_4125_cast_fp16_9)[name = string("transpose_3572")]; tensor reshape_651_cast_fp16 = reshape(shape = concat_2174, x = transpose_434_cast_fp16)[name = string("reshape_651_cast_fp16")]; tensor transpose_435_perm_0 = const()[name = string("transpose_435_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2175 = const()[name = string("concat_2175"), val = tensor([1, 64, 104])]; tensor transpose_435_cast_fp16 = transpose(perm = transpose_435_perm_0, x = var_4142_cast_fp16_9)[name = string("transpose_3571")]; tensor reshape_652_cast_fp16 = reshape(shape = concat_2175, x = transpose_435_cast_fp16)[name = string("reshape_652_cast_fp16")]; bool matmul_217_transpose_x_0 = const()[name = string("matmul_217_transpose_x_0"), val = bool(false)]; bool matmul_217_transpose_y_0 = const()[name = string("matmul_217_transpose_y_0"), val = bool(false)]; tensor matmul_217_cast_fp16 = matmul(transpose_x = matmul_217_transpose_x_0, transpose_y = matmul_217_transpose_y_0, x = reshape_651_cast_fp16, y = reshape_652_cast_fp16)[name = string("matmul_217_cast_fp16")]; tensor concat_2179 = const()[name = string("concat_2179"), val = tensor([1, 1, 104, 104])]; tensor reshape_653_cast_fp16 = reshape(shape = concat_2179, x = matmul_217_cast_fp16)[name = string("reshape_653_cast_fp16")]; tensor transpose_2905_perm_0 = const()[name = string("transpose_2905_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2905 = transpose(perm = transpose_2905_perm_0, x = reshape_653_cast_fp16)[name = string("transpose_3570")]; tensor w_871_cast_fp16 = add(x = transpose_2905, y = transpose_2305)[name = string("w_871_cast_fp16")]; tensor var_4253_cast_fp16 = softmax(axis = var_4069, x = w_871_cast_fp16)[name = string("op_4253_cast_fp16")]; string var_4255_equation_0 = const()[name = string("op_4255_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4255_cast_fp16 = einsum(equation = var_4255_equation_0, values = (var_4159_cast_fp16_9, var_4253_cast_fp16))[name = string("op_4255_cast_fp16")]; tensor transpose_436_perm_0 = const()[name = string("transpose_436_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2184 = const()[name = string("concat_2184"), val = tensor([1, 104, 64])]; tensor transpose_436_cast_fp16 = transpose(perm = transpose_436_perm_0, x = var_4125_cast_fp16_10)[name = string("transpose_3569")]; tensor reshape_654_cast_fp16 = reshape(shape = concat_2184, x = transpose_436_cast_fp16)[name = string("reshape_654_cast_fp16")]; tensor transpose_437_perm_0 = const()[name = string("transpose_437_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2185 = const()[name = string("concat_2185"), val = tensor([1, 64, 104])]; tensor transpose_437_cast_fp16 = transpose(perm = transpose_437_perm_0, x = var_4142_cast_fp16_10)[name = string("transpose_3568")]; tensor reshape_655_cast_fp16 = reshape(shape = concat_2185, x = transpose_437_cast_fp16)[name = string("reshape_655_cast_fp16")]; bool matmul_218_transpose_x_0 = const()[name = string("matmul_218_transpose_x_0"), val = bool(false)]; bool matmul_218_transpose_y_0 = const()[name = string("matmul_218_transpose_y_0"), val = bool(false)]; tensor matmul_218_cast_fp16 = matmul(transpose_x = matmul_218_transpose_x_0, transpose_y = matmul_218_transpose_y_0, x = reshape_654_cast_fp16, y = reshape_655_cast_fp16)[name = string("matmul_218_cast_fp16")]; tensor concat_2189 = const()[name = string("concat_2189"), val = tensor([1, 1, 104, 104])]; tensor reshape_656_cast_fp16 = reshape(shape = concat_2189, x = matmul_218_cast_fp16)[name = string("reshape_656_cast_fp16")]; tensor transpose_2906_perm_0 = const()[name = string("transpose_2906_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2906 = transpose(perm = transpose_2906_perm_0, x = reshape_656_cast_fp16)[name = string("transpose_3567")]; tensor w_875_cast_fp16 = add(x = transpose_2906, y = transpose_2305)[name = string("w_875_cast_fp16")]; tensor var_4261_cast_fp16 = softmax(axis = var_4069, x = w_875_cast_fp16)[name = string("op_4261_cast_fp16")]; string var_4263_equation_0 = const()[name = string("op_4263_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4263_cast_fp16 = einsum(equation = var_4263_equation_0, values = (var_4159_cast_fp16_10, var_4261_cast_fp16))[name = string("op_4263_cast_fp16")]; tensor transpose_438_perm_0 = const()[name = string("transpose_438_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2194 = const()[name = string("concat_2194"), val = tensor([1, 104, 64])]; tensor transpose_438_cast_fp16 = transpose(perm = transpose_438_perm_0, x = var_4125_cast_fp16_11)[name = string("transpose_3566")]; tensor reshape_657_cast_fp16 = reshape(shape = concat_2194, x = transpose_438_cast_fp16)[name = string("reshape_657_cast_fp16")]; tensor transpose_439_perm_0 = const()[name = string("transpose_439_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2195 = const()[name = string("concat_2195"), val = tensor([1, 64, 104])]; tensor transpose_439_cast_fp16 = transpose(perm = transpose_439_perm_0, x = var_4142_cast_fp16_11)[name = string("transpose_3565")]; tensor reshape_658_cast_fp16 = reshape(shape = concat_2195, x = transpose_439_cast_fp16)[name = string("reshape_658_cast_fp16")]; bool matmul_219_transpose_x_0 = const()[name = string("matmul_219_transpose_x_0"), val = bool(false)]; bool matmul_219_transpose_y_0 = const()[name = string("matmul_219_transpose_y_0"), val = bool(false)]; tensor matmul_219_cast_fp16 = matmul(transpose_x = matmul_219_transpose_x_0, transpose_y = matmul_219_transpose_y_0, x = reshape_657_cast_fp16, y = reshape_658_cast_fp16)[name = string("matmul_219_cast_fp16")]; tensor concat_2199 = const()[name = string("concat_2199"), val = tensor([1, 1, 104, 104])]; tensor reshape_659_cast_fp16 = reshape(shape = concat_2199, x = matmul_219_cast_fp16)[name = string("reshape_659_cast_fp16")]; tensor transpose_2907_perm_0 = const()[name = string("transpose_2907_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2907 = transpose(perm = transpose_2907_perm_0, x = reshape_659_cast_fp16)[name = string("transpose_3564")]; tensor w_879_cast_fp16 = add(x = transpose_2907, y = transpose_2305)[name = string("w_879_cast_fp16")]; tensor var_4269_cast_fp16 = softmax(axis = var_4069, x = w_879_cast_fp16)[name = string("op_4269_cast_fp16")]; string var_4271_equation_0 = const()[name = string("op_4271_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4271_cast_fp16 = einsum(equation = var_4271_equation_0, values = (var_4159_cast_fp16_11, var_4269_cast_fp16))[name = string("op_4271_cast_fp16")]; tensor transpose_440_perm_0 = const()[name = string("transpose_440_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2204 = const()[name = string("concat_2204"), val = tensor([1, 104, 64])]; tensor transpose_440_cast_fp16 = transpose(perm = transpose_440_perm_0, x = var_4125_cast_fp16_12)[name = string("transpose_3563")]; tensor reshape_660_cast_fp16 = reshape(shape = concat_2204, x = transpose_440_cast_fp16)[name = string("reshape_660_cast_fp16")]; tensor transpose_441_perm_0 = const()[name = string("transpose_441_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2205 = const()[name = string("concat_2205"), val = tensor([1, 64, 104])]; tensor transpose_441_cast_fp16 = transpose(perm = transpose_441_perm_0, x = var_4142_cast_fp16_12)[name = string("transpose_3562")]; tensor reshape_661_cast_fp16 = reshape(shape = concat_2205, x = transpose_441_cast_fp16)[name = string("reshape_661_cast_fp16")]; bool matmul_220_transpose_x_0 = const()[name = string("matmul_220_transpose_x_0"), val = bool(false)]; bool matmul_220_transpose_y_0 = const()[name = string("matmul_220_transpose_y_0"), val = bool(false)]; tensor matmul_220_cast_fp16 = matmul(transpose_x = matmul_220_transpose_x_0, transpose_y = matmul_220_transpose_y_0, x = reshape_660_cast_fp16, y = reshape_661_cast_fp16)[name = string("matmul_220_cast_fp16")]; tensor concat_2209 = const()[name = string("concat_2209"), val = tensor([1, 1, 104, 104])]; tensor reshape_662_cast_fp16 = reshape(shape = concat_2209, x = matmul_220_cast_fp16)[name = string("reshape_662_cast_fp16")]; tensor transpose_2908_perm_0 = const()[name = string("transpose_2908_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2908 = transpose(perm = transpose_2908_perm_0, x = reshape_662_cast_fp16)[name = string("transpose_3561")]; tensor w_883_cast_fp16 = add(x = transpose_2908, y = transpose_2305)[name = string("w_883_cast_fp16")]; tensor var_4277_cast_fp16 = softmax(axis = var_4069, x = w_883_cast_fp16)[name = string("op_4277_cast_fp16")]; string var_4279_equation_0 = const()[name = string("op_4279_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4279_cast_fp16 = einsum(equation = var_4279_equation_0, values = (var_4159_cast_fp16_12, var_4277_cast_fp16))[name = string("op_4279_cast_fp16")]; tensor transpose_442_perm_0 = const()[name = string("transpose_442_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2214 = const()[name = string("concat_2214"), val = tensor([1, 104, 64])]; tensor transpose_442_cast_fp16 = transpose(perm = transpose_442_perm_0, x = var_4125_cast_fp16_13)[name = string("transpose_3560")]; tensor reshape_663_cast_fp16 = reshape(shape = concat_2214, x = transpose_442_cast_fp16)[name = string("reshape_663_cast_fp16")]; tensor transpose_443_perm_0 = const()[name = string("transpose_443_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2215 = const()[name = string("concat_2215"), val = tensor([1, 64, 104])]; tensor transpose_443_cast_fp16 = transpose(perm = transpose_443_perm_0, x = var_4142_cast_fp16_13)[name = string("transpose_3559")]; tensor reshape_664_cast_fp16 = reshape(shape = concat_2215, x = transpose_443_cast_fp16)[name = string("reshape_664_cast_fp16")]; bool matmul_221_transpose_x_0 = const()[name = string("matmul_221_transpose_x_0"), val = bool(false)]; bool matmul_221_transpose_y_0 = const()[name = string("matmul_221_transpose_y_0"), val = bool(false)]; tensor matmul_221_cast_fp16 = matmul(transpose_x = matmul_221_transpose_x_0, transpose_y = matmul_221_transpose_y_0, x = reshape_663_cast_fp16, y = reshape_664_cast_fp16)[name = string("matmul_221_cast_fp16")]; tensor concat_2219 = const()[name = string("concat_2219"), val = tensor([1, 1, 104, 104])]; tensor reshape_665_cast_fp16 = reshape(shape = concat_2219, x = matmul_221_cast_fp16)[name = string("reshape_665_cast_fp16")]; tensor transpose_2909_perm_0 = const()[name = string("transpose_2909_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2909 = transpose(perm = transpose_2909_perm_0, x = reshape_665_cast_fp16)[name = string("transpose_3558")]; tensor w_887_cast_fp16 = add(x = transpose_2909, y = transpose_2305)[name = string("w_887_cast_fp16")]; tensor var_4285_cast_fp16 = softmax(axis = var_4069, x = w_887_cast_fp16)[name = string("op_4285_cast_fp16")]; string var_4287_equation_0 = const()[name = string("op_4287_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4287_cast_fp16 = einsum(equation = var_4287_equation_0, values = (var_4159_cast_fp16_13, var_4285_cast_fp16))[name = string("op_4287_cast_fp16")]; tensor transpose_444_perm_0 = const()[name = string("transpose_444_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2224 = const()[name = string("concat_2224"), val = tensor([1, 104, 64])]; tensor transpose_444_cast_fp16 = transpose(perm = transpose_444_perm_0, x = var_4125_cast_fp16_14)[name = string("transpose_3557")]; tensor reshape_666_cast_fp16 = reshape(shape = concat_2224, x = transpose_444_cast_fp16)[name = string("reshape_666_cast_fp16")]; tensor transpose_445_perm_0 = const()[name = string("transpose_445_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2225 = const()[name = string("concat_2225"), val = tensor([1, 64, 104])]; tensor transpose_445_cast_fp16 = transpose(perm = transpose_445_perm_0, x = var_4142_cast_fp16_14)[name = string("transpose_3556")]; tensor reshape_667_cast_fp16 = reshape(shape = concat_2225, x = transpose_445_cast_fp16)[name = string("reshape_667_cast_fp16")]; bool matmul_222_transpose_x_0 = const()[name = string("matmul_222_transpose_x_0"), val = bool(false)]; bool matmul_222_transpose_y_0 = const()[name = string("matmul_222_transpose_y_0"), val = bool(false)]; tensor matmul_222_cast_fp16 = matmul(transpose_x = matmul_222_transpose_x_0, transpose_y = matmul_222_transpose_y_0, x = reshape_666_cast_fp16, y = reshape_667_cast_fp16)[name = string("matmul_222_cast_fp16")]; tensor concat_2229 = const()[name = string("concat_2229"), val = tensor([1, 1, 104, 104])]; tensor reshape_668_cast_fp16 = reshape(shape = concat_2229, x = matmul_222_cast_fp16)[name = string("reshape_668_cast_fp16")]; tensor transpose_2910_perm_0 = const()[name = string("transpose_2910_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2910 = transpose(perm = transpose_2910_perm_0, x = reshape_668_cast_fp16)[name = string("transpose_3555")]; tensor w_891_cast_fp16 = add(x = transpose_2910, y = transpose_2305)[name = string("w_891_cast_fp16")]; tensor var_4293_cast_fp16 = softmax(axis = var_4069, x = w_891_cast_fp16)[name = string("op_4293_cast_fp16")]; string var_4295_equation_0 = const()[name = string("op_4295_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4295_cast_fp16 = einsum(equation = var_4295_equation_0, values = (var_4159_cast_fp16_14, var_4293_cast_fp16))[name = string("op_4295_cast_fp16")]; tensor transpose_446_perm_0 = const()[name = string("transpose_446_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2234 = const()[name = string("concat_2234"), val = tensor([1, 104, 64])]; tensor transpose_446_cast_fp16 = transpose(perm = transpose_446_perm_0, x = var_4125_cast_fp16_15)[name = string("transpose_3554")]; tensor reshape_669_cast_fp16 = reshape(shape = concat_2234, x = transpose_446_cast_fp16)[name = string("reshape_669_cast_fp16")]; tensor transpose_447_perm_0 = const()[name = string("transpose_447_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2235 = const()[name = string("concat_2235"), val = tensor([1, 64, 104])]; tensor transpose_447_cast_fp16 = transpose(perm = transpose_447_perm_0, x = var_4142_cast_fp16_15)[name = string("transpose_3553")]; tensor reshape_670_cast_fp16 = reshape(shape = concat_2235, x = transpose_447_cast_fp16)[name = string("reshape_670_cast_fp16")]; bool matmul_223_transpose_x_0 = const()[name = string("matmul_223_transpose_x_0"), val = bool(false)]; bool matmul_223_transpose_y_0 = const()[name = string("matmul_223_transpose_y_0"), val = bool(false)]; tensor matmul_223_cast_fp16 = matmul(transpose_x = matmul_223_transpose_x_0, transpose_y = matmul_223_transpose_y_0, x = reshape_669_cast_fp16, y = reshape_670_cast_fp16)[name = string("matmul_223_cast_fp16")]; tensor concat_2239 = const()[name = string("concat_2239"), val = tensor([1, 1, 104, 104])]; tensor reshape_671_cast_fp16 = reshape(shape = concat_2239, x = matmul_223_cast_fp16)[name = string("reshape_671_cast_fp16")]; tensor transpose_2911_perm_0 = const()[name = string("transpose_2911_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2911 = transpose(perm = transpose_2911_perm_0, x = reshape_671_cast_fp16)[name = string("transpose_3552")]; tensor w_895_cast_fp16 = add(x = transpose_2911, y = transpose_2305)[name = string("w_895_cast_fp16")]; tensor var_4301_cast_fp16 = softmax(axis = var_4069, x = w_895_cast_fp16)[name = string("op_4301_cast_fp16")]; string var_4303_equation_0 = const()[name = string("op_4303_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4303_cast_fp16 = einsum(equation = var_4303_equation_0, values = (var_4159_cast_fp16_15, var_4301_cast_fp16))[name = string("op_4303_cast_fp16")]; bool input_115_interleave_0 = const()[name = string("input_115_interleave_0"), val = bool(false)]; tensor input_115_cast_fp16 = concat(axis = var_4069, interleave = input_115_interleave_0, values = (var_4183_cast_fp16, var_4191_cast_fp16, var_4199_cast_fp16, var_4207_cast_fp16, var_4215_cast_fp16, var_4223_cast_fp16, var_4231_cast_fp16, var_4239_cast_fp16, var_4247_cast_fp16, var_4255_cast_fp16, var_4263_cast_fp16, var_4271_cast_fp16, var_4279_cast_fp16, var_4287_cast_fp16, var_4295_cast_fp16, var_4303_cast_fp16))[name = string("input_115_cast_fp16")]; string var_4312_pad_type_0 = const()[name = string("op_4312_pad_type_0"), val = string("valid")]; tensor var_4312_strides_0 = const()[name = string("op_4312_strides_0"), val = tensor([1, 1])]; tensor var_4312_pad_0 = const()[name = string("op_4312_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4312_dilations_0 = const()[name = string("op_4312_dilations_0"), val = tensor([1, 1])]; int32 var_4312_groups_0 = const()[name = string("op_4312_groups_0"), val = int32(1)]; tensor layers_13_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_13_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(357888704)))]; tensor layers_13_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_13_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(359985920)))]; tensor var_4312_cast_fp16 = conv(bias = layers_13_self_attn_out_proj_bias_to_fp16, dilations = var_4312_dilations_0, groups = var_4312_groups_0, pad = var_4312_pad_0, pad_type = var_4312_pad_type_0, strides = var_4312_strides_0, weight = layers_13_self_attn_out_proj_weight_to_fp16, x = input_115_cast_fp16)[name = string("op_4312_cast_fp16")]; tensor x_147_cast_fp16 = add(x = x_143_cast_fp16, y = var_4312_cast_fp16)[name = string("x_147_cast_fp16")]; tensor mu_55_axes_0 = const()[name = string("mu_55_axes_0"), val = tensor([1])]; bool mu_55_keep_dims_0 = const()[name = string("mu_55_keep_dims_0"), val = bool(true)]; tensor mu_55_cast_fp16 = reduce_mean(axes = mu_55_axes_0, keep_dims = mu_55_keep_dims_0, x = x_147_cast_fp16)[name = string("mu_55_cast_fp16")]; tensor var_4318_cast_fp16 = sub(x = x_147_cast_fp16, y = mu_55_cast_fp16)[name = string("op_4318_cast_fp16")]; fp16 var_4072_promoted_1_to_fp16 = const()[name = string("op_4072_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_4319_cast_fp16 = pow(x = var_4318_cast_fp16, y = var_4072_promoted_1_to_fp16)[name = string("op_4319_cast_fp16")]; tensor var_55_axes_0 = const()[name = string("var_55_axes_0"), val = tensor([1])]; bool var_55_keep_dims_0 = const()[name = string("var_55_keep_dims_0"), val = bool(true)]; tensor var_55_cast_fp16 = reduce_mean(axes = var_55_axes_0, keep_dims = var_55_keep_dims_0, x = var_4319_cast_fp16)[name = string("var_55_cast_fp16")]; fp16 var_4323_to_fp16 = const()[name = string("op_4323_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4324_cast_fp16 = add(x = var_55_cast_fp16, y = var_4323_to_fp16)[name = string("op_4324_cast_fp16")]; fp32 var_4325_epsilon_0 = const()[name = string("op_4325_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4325_cast_fp16 = rsqrt(epsilon = var_4325_epsilon_0, x = var_4324_cast_fp16)[name = string("op_4325_cast_fp16")]; tensor x_149_cast_fp16 = mul(x = var_4318_cast_fp16, y = var_4325_cast_fp16)[name = string("x_149_cast_fp16")]; tensor input_117_gamma_0_to_fp16 = const()[name = string("input_117_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(359988032)))]; tensor input_117_beta_0_to_fp16 = const()[name = string("input_117_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(359990144)))]; fp16 input_117_epsilon_0_to_fp16 = const()[name = string("input_117_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_117_cast_fp16 = batch_norm(beta = input_117_beta_0_to_fp16, epsilon = input_117_epsilon_0_to_fp16, gamma = input_117_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_149_cast_fp16)[name = string("input_117_cast_fp16")]; string x_151_pad_type_0 = const()[name = string("x_151_pad_type_0"), val = string("valid")]; tensor x_151_strides_0 = const()[name = string("x_151_strides_0"), val = tensor([1, 1])]; tensor x_151_pad_0 = const()[name = string("x_151_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_151_dilations_0 = const()[name = string("x_151_dilations_0"), val = tensor([1, 1])]; int32 x_151_groups_0 = const()[name = string("x_151_groups_0"), val = int32(1)]; tensor layers_13_fc1_weight_to_fp16 = const()[name = string("layers_13_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(359992256)))]; tensor layers_13_fc1_bias_to_fp16 = const()[name = string("layers_13_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(368380928)))]; tensor x_151_cast_fp16 = conv(bias = layers_13_fc1_bias_to_fp16, dilations = x_151_dilations_0, groups = x_151_groups_0, pad = x_151_pad_0, pad_type = x_151_pad_type_0, strides = x_151_strides_0, weight = layers_13_fc1_weight_to_fp16, x = input_117_cast_fp16)[name = string("x_151_cast_fp16")]; fp16 var_4340_to_fp16 = const()[name = string("op_4340_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_4341_cast_fp16 = mul(x = x_151_cast_fp16, y = var_4340_to_fp16)[name = string("op_4341_cast_fp16")]; tensor var_4342_cast_fp16 = mul(x = var_4341_cast_fp16, y = x_151_cast_fp16)[name = string("op_4342_cast_fp16")]; tensor var_4343_cast_fp16 = mul(x = var_4342_cast_fp16, y = x_151_cast_fp16)[name = string("op_4343_cast_fp16")]; tensor var_4344_cast_fp16 = add(x = x_151_cast_fp16, y = var_4343_cast_fp16)[name = string("op_4344_cast_fp16")]; fp16 var_4345_to_fp16 = const()[name = string("op_4345_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_33_cast_fp16 = mul(x = var_4344_cast_fp16, y = var_4345_to_fp16)[name = string("u_33_cast_fp16")]; fp16 var_4347_to_fp16 = const()[name = string("op_4347_to_fp16"), val = fp16(0x1p-1)]; tensor var_4348_cast_fp16 = mul(x = x_151_cast_fp16, y = var_4347_to_fp16)[name = string("op_4348_cast_fp16")]; tensor var_4349_cast_fp16 = tanh(x = u_33_cast_fp16)[name = string("op_4349_cast_fp16")]; fp16 var_4350_to_fp16 = const()[name = string("op_4350_to_fp16"), val = fp16(0x1p+0)]; tensor var_4351_cast_fp16 = add(x = var_4349_cast_fp16, y = var_4350_to_fp16)[name = string("op_4351_cast_fp16")]; tensor input_119_cast_fp16 = mul(x = var_4348_cast_fp16, y = var_4351_cast_fp16)[name = string("input_119_cast_fp16")]; string h_27_pad_type_0 = const()[name = string("h_27_pad_type_0"), val = string("valid")]; tensor h_27_strides_0 = const()[name = string("h_27_strides_0"), val = tensor([1, 1])]; tensor h_27_pad_0 = const()[name = string("h_27_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_27_dilations_0 = const()[name = string("h_27_dilations_0"), val = tensor([1, 1])]; int32 h_27_groups_0 = const()[name = string("h_27_groups_0"), val = int32(1)]; tensor layers_13_fc2_weight_to_fp16 = const()[name = string("layers_13_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(368389184)))]; tensor layers_13_fc2_bias_to_fp16 = const()[name = string("layers_13_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(376777856)))]; tensor h_27_cast_fp16 = conv(bias = layers_13_fc2_bias_to_fp16, dilations = h_27_dilations_0, groups = h_27_groups_0, pad = h_27_pad_0, pad_type = h_27_pad_type_0, strides = h_27_strides_0, weight = layers_13_fc2_weight_to_fp16, x = input_119_cast_fp16)[name = string("h_27_cast_fp16")]; tensor x_153_cast_fp16 = add(x = x_147_cast_fp16, y = h_27_cast_fp16)[name = string("x_153_cast_fp16")]; int32 var_4367 = const()[name = string("op_4367"), val = int32(1)]; tensor mu_57_axes_0 = const()[name = string("mu_57_axes_0"), val = tensor([1])]; bool mu_57_keep_dims_0 = const()[name = string("mu_57_keep_dims_0"), val = bool(true)]; tensor mu_57_cast_fp16 = reduce_mean(axes = mu_57_axes_0, keep_dims = mu_57_keep_dims_0, x = x_153_cast_fp16)[name = string("mu_57_cast_fp16")]; tensor var_4381_cast_fp16 = sub(x = x_153_cast_fp16, y = mu_57_cast_fp16)[name = string("op_4381_cast_fp16")]; fp16 var_4370_promoted_to_fp16 = const()[name = string("op_4370_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_4382_cast_fp16 = pow(x = var_4381_cast_fp16, y = var_4370_promoted_to_fp16)[name = string("op_4382_cast_fp16")]; tensor var_57_axes_0 = const()[name = string("var_57_axes_0"), val = tensor([1])]; bool var_57_keep_dims_0 = const()[name = string("var_57_keep_dims_0"), val = bool(true)]; tensor var_57_cast_fp16 = reduce_mean(axes = var_57_axes_0, keep_dims = var_57_keep_dims_0, x = var_4382_cast_fp16)[name = string("var_57_cast_fp16")]; fp16 var_4386_to_fp16 = const()[name = string("op_4386_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4387_cast_fp16 = add(x = var_57_cast_fp16, y = var_4386_to_fp16)[name = string("op_4387_cast_fp16")]; fp32 var_4388_epsilon_0 = const()[name = string("op_4388_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4388_cast_fp16 = rsqrt(epsilon = var_4388_epsilon_0, x = var_4387_cast_fp16)[name = string("op_4388_cast_fp16")]; tensor x_155_cast_fp16 = mul(x = var_4381_cast_fp16, y = var_4388_cast_fp16)[name = string("x_155_cast_fp16")]; tensor input_121_gamma_0_to_fp16 = const()[name = string("input_121_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(376779968)))]; tensor input_121_beta_0_to_fp16 = const()[name = string("input_121_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(376782080)))]; fp16 input_121_epsilon_0_to_fp16 = const()[name = string("input_121_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_121_cast_fp16 = batch_norm(beta = input_121_beta_0_to_fp16, epsilon = input_121_epsilon_0_to_fp16, gamma = input_121_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_155_cast_fp16)[name = string("input_121_cast_fp16")]; string var_4406_pad_type_0 = const()[name = string("op_4406_pad_type_0"), val = string("valid")]; tensor var_4406_strides_0 = const()[name = string("op_4406_strides_0"), val = tensor([1, 1])]; tensor var_4406_pad_0 = const()[name = string("op_4406_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4406_dilations_0 = const()[name = string("op_4406_dilations_0"), val = tensor([1, 1])]; int32 var_4406_groups_0 = const()[name = string("op_4406_groups_0"), val = int32(1)]; tensor var_4408_weight_0_to_fp16 = const()[name = string("op_4408_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(376784192)))]; tensor var_4408_bias_0_to_fp16 = const()[name = string("op_4408_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(378881408)))]; tensor var_4408_cast_fp16 = conv(bias = var_4408_bias_0_to_fp16, dilations = var_4406_dilations_0, groups = var_4406_groups_0, pad = var_4406_pad_0, pad_type = var_4406_pad_type_0, strides = var_4406_strides_0, weight = var_4408_weight_0_to_fp16, x = input_121_cast_fp16)[name = string("op_4408_cast_fp16")]; string var_4415_pad_type_0 = const()[name = string("op_4415_pad_type_0"), val = string("valid")]; tensor var_4415_strides_0 = const()[name = string("op_4415_strides_0"), val = tensor([1, 1])]; tensor var_4415_pad_0 = const()[name = string("op_4415_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4415_dilations_0 = const()[name = string("op_4415_dilations_0"), val = tensor([1, 1])]; int32 var_4415_groups_0 = const()[name = string("op_4415_groups_0"), val = int32(1)]; tensor layers_14_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_14_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(378883520)))]; tensor layers_14_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_14_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(380980736)))]; tensor var_4415_cast_fp16 = conv(bias = layers_14_self_attn_k_proj_bias_to_fp16, dilations = var_4415_dilations_0, groups = var_4415_groups_0, pad = var_4415_pad_0, pad_type = var_4415_pad_type_0, strides = var_4415_strides_0, weight = layers_14_self_attn_k_proj_weight_to_fp16, x = input_121_cast_fp16)[name = string("op_4415_cast_fp16")]; string var_4422_pad_type_0 = const()[name = string("op_4422_pad_type_0"), val = string("valid")]; tensor var_4422_strides_0 = const()[name = string("op_4422_strides_0"), val = tensor([1, 1])]; tensor var_4422_pad_0 = const()[name = string("op_4422_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4422_dilations_0 = const()[name = string("op_4422_dilations_0"), val = tensor([1, 1])]; int32 var_4422_groups_0 = const()[name = string("op_4422_groups_0"), val = int32(1)]; tensor layers_14_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_14_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(380982848)))]; tensor layers_14_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_14_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(383080064)))]; tensor var_4422_cast_fp16 = conv(bias = layers_14_self_attn_v_proj_bias_to_fp16, dilations = var_4422_dilations_0, groups = var_4422_groups_0, pad = var_4422_pad_0, pad_type = var_4422_pad_type_0, strides = var_4422_strides_0, weight = layers_14_self_attn_v_proj_weight_to_fp16, x = input_121_cast_fp16)[name = string("op_4422_cast_fp16")]; tensor tile_42 = const()[name = string("tile_42"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(383082176)))]; int32 var_4423_axis_0 = const()[name = string("op_4423_axis_0"), val = int32(1)]; tensor var_4423_cast_fp16_0, tensor var_4423_cast_fp16_1, tensor var_4423_cast_fp16_2, tensor var_4423_cast_fp16_3, tensor var_4423_cast_fp16_4, tensor var_4423_cast_fp16_5, tensor var_4423_cast_fp16_6, tensor var_4423_cast_fp16_7, tensor var_4423_cast_fp16_8, tensor var_4423_cast_fp16_9, tensor var_4423_cast_fp16_10, tensor var_4423_cast_fp16_11, tensor var_4423_cast_fp16_12, tensor var_4423_cast_fp16_13, tensor var_4423_cast_fp16_14, tensor var_4423_cast_fp16_15 = split(axis = var_4423_axis_0, split_sizes = tile_42, x = var_4408_cast_fp16)[name = string("op_4423_cast_fp16")]; tensor tile_43 = const()[name = string("tile_43"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(383082304)))]; int32 var_4440_axis_0 = const()[name = string("op_4440_axis_0"), val = int32(1)]; tensor var_4440_cast_fp16_0, tensor var_4440_cast_fp16_1, tensor var_4440_cast_fp16_2, tensor var_4440_cast_fp16_3, tensor var_4440_cast_fp16_4, tensor var_4440_cast_fp16_5, tensor var_4440_cast_fp16_6, tensor var_4440_cast_fp16_7, tensor var_4440_cast_fp16_8, tensor var_4440_cast_fp16_9, tensor var_4440_cast_fp16_10, tensor var_4440_cast_fp16_11, tensor var_4440_cast_fp16_12, tensor var_4440_cast_fp16_13, tensor var_4440_cast_fp16_14, tensor var_4440_cast_fp16_15 = split(axis = var_4440_axis_0, split_sizes = tile_43, x = var_4415_cast_fp16)[name = string("op_4440_cast_fp16")]; tensor tile_44 = const()[name = string("tile_44"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(383082432)))]; int32 var_4457_axis_0 = const()[name = string("op_4457_axis_0"), val = int32(1)]; tensor var_4457_cast_fp16_0, tensor var_4457_cast_fp16_1, tensor var_4457_cast_fp16_2, tensor var_4457_cast_fp16_3, tensor var_4457_cast_fp16_4, tensor var_4457_cast_fp16_5, tensor var_4457_cast_fp16_6, tensor var_4457_cast_fp16_7, tensor var_4457_cast_fp16_8, tensor var_4457_cast_fp16_9, tensor var_4457_cast_fp16_10, tensor var_4457_cast_fp16_11, tensor var_4457_cast_fp16_12, tensor var_4457_cast_fp16_13, tensor var_4457_cast_fp16_14, tensor var_4457_cast_fp16_15 = split(axis = var_4457_axis_0, split_sizes = tile_44, x = var_4422_cast_fp16)[name = string("op_4457_cast_fp16")]; tensor transpose_448_perm_0 = const()[name = string("transpose_448_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2244 = const()[name = string("concat_2244"), val = tensor([1, 104, 64])]; tensor transpose_448_cast_fp16 = transpose(perm = transpose_448_perm_0, x = var_4423_cast_fp16_0)[name = string("transpose_3551")]; tensor reshape_672_cast_fp16 = reshape(shape = concat_2244, x = transpose_448_cast_fp16)[name = string("reshape_672_cast_fp16")]; tensor transpose_449_perm_0 = const()[name = string("transpose_449_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2245 = const()[name = string("concat_2245"), val = tensor([1, 64, 104])]; tensor transpose_449_cast_fp16 = transpose(perm = transpose_449_perm_0, x = var_4440_cast_fp16_0)[name = string("transpose_3550")]; tensor reshape_673_cast_fp16 = reshape(shape = concat_2245, x = transpose_449_cast_fp16)[name = string("reshape_673_cast_fp16")]; bool matmul_224_transpose_x_0 = const()[name = string("matmul_224_transpose_x_0"), val = bool(false)]; bool matmul_224_transpose_y_0 = const()[name = string("matmul_224_transpose_y_0"), val = bool(false)]; tensor matmul_224_cast_fp16 = matmul(transpose_x = matmul_224_transpose_x_0, transpose_y = matmul_224_transpose_y_0, x = reshape_672_cast_fp16, y = reshape_673_cast_fp16)[name = string("matmul_224_cast_fp16")]; tensor concat_2249 = const()[name = string("concat_2249"), val = tensor([1, 1, 104, 104])]; tensor reshape_674_cast_fp16 = reshape(shape = concat_2249, x = matmul_224_cast_fp16)[name = string("reshape_674_cast_fp16")]; tensor transpose_2912_perm_0 = const()[name = string("transpose_2912_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2912 = transpose(perm = transpose_2912_perm_0, x = reshape_674_cast_fp16)[name = string("transpose_3549")]; tensor w_899_cast_fp16 = add(x = transpose_2912, y = transpose_2305)[name = string("w_899_cast_fp16")]; tensor var_4479_cast_fp16 = softmax(axis = var_4367, x = w_899_cast_fp16)[name = string("op_4479_cast_fp16")]; string var_4481_equation_0 = const()[name = string("op_4481_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4481_cast_fp16 = einsum(equation = var_4481_equation_0, values = (var_4457_cast_fp16_0, var_4479_cast_fp16))[name = string("op_4481_cast_fp16")]; tensor transpose_450_perm_0 = const()[name = string("transpose_450_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2254 = const()[name = string("concat_2254"), val = tensor([1, 104, 64])]; tensor transpose_450_cast_fp16 = transpose(perm = transpose_450_perm_0, x = var_4423_cast_fp16_1)[name = string("transpose_3548")]; tensor reshape_675_cast_fp16 = reshape(shape = concat_2254, x = transpose_450_cast_fp16)[name = string("reshape_675_cast_fp16")]; tensor transpose_451_perm_0 = const()[name = string("transpose_451_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2255 = const()[name = string("concat_2255"), val = tensor([1, 64, 104])]; tensor transpose_451_cast_fp16 = transpose(perm = transpose_451_perm_0, x = var_4440_cast_fp16_1)[name = string("transpose_3547")]; tensor reshape_676_cast_fp16 = reshape(shape = concat_2255, x = transpose_451_cast_fp16)[name = string("reshape_676_cast_fp16")]; bool matmul_225_transpose_x_0 = const()[name = string("matmul_225_transpose_x_0"), val = bool(false)]; bool matmul_225_transpose_y_0 = const()[name = string("matmul_225_transpose_y_0"), val = bool(false)]; tensor matmul_225_cast_fp16 = matmul(transpose_x = matmul_225_transpose_x_0, transpose_y = matmul_225_transpose_y_0, x = reshape_675_cast_fp16, y = reshape_676_cast_fp16)[name = string("matmul_225_cast_fp16")]; tensor concat_2259 = const()[name = string("concat_2259"), val = tensor([1, 1, 104, 104])]; tensor reshape_677_cast_fp16 = reshape(shape = concat_2259, x = matmul_225_cast_fp16)[name = string("reshape_677_cast_fp16")]; tensor transpose_2913_perm_0 = const()[name = string("transpose_2913_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2913 = transpose(perm = transpose_2913_perm_0, x = reshape_677_cast_fp16)[name = string("transpose_3546")]; tensor w_903_cast_fp16 = add(x = transpose_2913, y = transpose_2305)[name = string("w_903_cast_fp16")]; tensor var_4487_cast_fp16 = softmax(axis = var_4367, x = w_903_cast_fp16)[name = string("op_4487_cast_fp16")]; string var_4489_equation_0 = const()[name = string("op_4489_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4489_cast_fp16 = einsum(equation = var_4489_equation_0, values = (var_4457_cast_fp16_1, var_4487_cast_fp16))[name = string("op_4489_cast_fp16")]; tensor transpose_452_perm_0 = const()[name = string("transpose_452_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2264 = const()[name = string("concat_2264"), val = tensor([1, 104, 64])]; tensor transpose_452_cast_fp16 = transpose(perm = transpose_452_perm_0, x = var_4423_cast_fp16_2)[name = string("transpose_3545")]; tensor reshape_678_cast_fp16 = reshape(shape = concat_2264, x = transpose_452_cast_fp16)[name = string("reshape_678_cast_fp16")]; tensor transpose_453_perm_0 = const()[name = string("transpose_453_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2265 = const()[name = string("concat_2265"), val = tensor([1, 64, 104])]; tensor transpose_453_cast_fp16 = transpose(perm = transpose_453_perm_0, x = var_4440_cast_fp16_2)[name = string("transpose_3544")]; tensor reshape_679_cast_fp16 = reshape(shape = concat_2265, x = transpose_453_cast_fp16)[name = string("reshape_679_cast_fp16")]; bool matmul_226_transpose_x_0 = const()[name = string("matmul_226_transpose_x_0"), val = bool(false)]; bool matmul_226_transpose_y_0 = const()[name = string("matmul_226_transpose_y_0"), val = bool(false)]; tensor matmul_226_cast_fp16 = matmul(transpose_x = matmul_226_transpose_x_0, transpose_y = matmul_226_transpose_y_0, x = reshape_678_cast_fp16, y = reshape_679_cast_fp16)[name = string("matmul_226_cast_fp16")]; tensor concat_2269 = const()[name = string("concat_2269"), val = tensor([1, 1, 104, 104])]; tensor reshape_680_cast_fp16 = reshape(shape = concat_2269, x = matmul_226_cast_fp16)[name = string("reshape_680_cast_fp16")]; tensor transpose_2914_perm_0 = const()[name = string("transpose_2914_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2914 = transpose(perm = transpose_2914_perm_0, x = reshape_680_cast_fp16)[name = string("transpose_3543")]; tensor w_907_cast_fp16 = add(x = transpose_2914, y = transpose_2305)[name = string("w_907_cast_fp16")]; tensor var_4495_cast_fp16 = softmax(axis = var_4367, x = w_907_cast_fp16)[name = string("op_4495_cast_fp16")]; string var_4497_equation_0 = const()[name = string("op_4497_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4497_cast_fp16 = einsum(equation = var_4497_equation_0, values = (var_4457_cast_fp16_2, var_4495_cast_fp16))[name = string("op_4497_cast_fp16")]; tensor transpose_454_perm_0 = const()[name = string("transpose_454_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2274 = const()[name = string("concat_2274"), val = tensor([1, 104, 64])]; tensor transpose_454_cast_fp16 = transpose(perm = transpose_454_perm_0, x = var_4423_cast_fp16_3)[name = string("transpose_3542")]; tensor reshape_681_cast_fp16 = reshape(shape = concat_2274, x = transpose_454_cast_fp16)[name = string("reshape_681_cast_fp16")]; tensor transpose_455_perm_0 = const()[name = string("transpose_455_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2275 = const()[name = string("concat_2275"), val = tensor([1, 64, 104])]; tensor transpose_455_cast_fp16 = transpose(perm = transpose_455_perm_0, x = var_4440_cast_fp16_3)[name = string("transpose_3541")]; tensor reshape_682_cast_fp16 = reshape(shape = concat_2275, x = transpose_455_cast_fp16)[name = string("reshape_682_cast_fp16")]; bool matmul_227_transpose_x_0 = const()[name = string("matmul_227_transpose_x_0"), val = bool(false)]; bool matmul_227_transpose_y_0 = const()[name = string("matmul_227_transpose_y_0"), val = bool(false)]; tensor matmul_227_cast_fp16 = matmul(transpose_x = matmul_227_transpose_x_0, transpose_y = matmul_227_transpose_y_0, x = reshape_681_cast_fp16, y = reshape_682_cast_fp16)[name = string("matmul_227_cast_fp16")]; tensor concat_2279 = const()[name = string("concat_2279"), val = tensor([1, 1, 104, 104])]; tensor reshape_683_cast_fp16 = reshape(shape = concat_2279, x = matmul_227_cast_fp16)[name = string("reshape_683_cast_fp16")]; tensor transpose_2915_perm_0 = const()[name = string("transpose_2915_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2915 = transpose(perm = transpose_2915_perm_0, x = reshape_683_cast_fp16)[name = string("transpose_3540")]; tensor w_911_cast_fp16 = add(x = transpose_2915, y = transpose_2305)[name = string("w_911_cast_fp16")]; tensor var_4503_cast_fp16 = softmax(axis = var_4367, x = w_911_cast_fp16)[name = string("op_4503_cast_fp16")]; string var_4505_equation_0 = const()[name = string("op_4505_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4505_cast_fp16 = einsum(equation = var_4505_equation_0, values = (var_4457_cast_fp16_3, var_4503_cast_fp16))[name = string("op_4505_cast_fp16")]; tensor transpose_456_perm_0 = const()[name = string("transpose_456_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2284 = const()[name = string("concat_2284"), val = tensor([1, 104, 64])]; tensor transpose_456_cast_fp16 = transpose(perm = transpose_456_perm_0, x = var_4423_cast_fp16_4)[name = string("transpose_3539")]; tensor reshape_684_cast_fp16 = reshape(shape = concat_2284, x = transpose_456_cast_fp16)[name = string("reshape_684_cast_fp16")]; tensor transpose_457_perm_0 = const()[name = string("transpose_457_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2285 = const()[name = string("concat_2285"), val = tensor([1, 64, 104])]; tensor transpose_457_cast_fp16 = transpose(perm = transpose_457_perm_0, x = var_4440_cast_fp16_4)[name = string("transpose_3538")]; tensor reshape_685_cast_fp16 = reshape(shape = concat_2285, x = transpose_457_cast_fp16)[name = string("reshape_685_cast_fp16")]; bool matmul_228_transpose_x_0 = const()[name = string("matmul_228_transpose_x_0"), val = bool(false)]; bool matmul_228_transpose_y_0 = const()[name = string("matmul_228_transpose_y_0"), val = bool(false)]; tensor matmul_228_cast_fp16 = matmul(transpose_x = matmul_228_transpose_x_0, transpose_y = matmul_228_transpose_y_0, x = reshape_684_cast_fp16, y = reshape_685_cast_fp16)[name = string("matmul_228_cast_fp16")]; tensor concat_2289 = const()[name = string("concat_2289"), val = tensor([1, 1, 104, 104])]; tensor reshape_686_cast_fp16 = reshape(shape = concat_2289, x = matmul_228_cast_fp16)[name = string("reshape_686_cast_fp16")]; tensor transpose_2916_perm_0 = const()[name = string("transpose_2916_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2916 = transpose(perm = transpose_2916_perm_0, x = reshape_686_cast_fp16)[name = string("transpose_3537")]; tensor w_915_cast_fp16 = add(x = transpose_2916, y = transpose_2305)[name = string("w_915_cast_fp16")]; tensor var_4511_cast_fp16 = softmax(axis = var_4367, x = w_915_cast_fp16)[name = string("op_4511_cast_fp16")]; string var_4513_equation_0 = const()[name = string("op_4513_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4513_cast_fp16 = einsum(equation = var_4513_equation_0, values = (var_4457_cast_fp16_4, var_4511_cast_fp16))[name = string("op_4513_cast_fp16")]; tensor transpose_458_perm_0 = const()[name = string("transpose_458_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2294 = const()[name = string("concat_2294"), val = tensor([1, 104, 64])]; tensor transpose_458_cast_fp16 = transpose(perm = transpose_458_perm_0, x = var_4423_cast_fp16_5)[name = string("transpose_3536")]; tensor reshape_687_cast_fp16 = reshape(shape = concat_2294, x = transpose_458_cast_fp16)[name = string("reshape_687_cast_fp16")]; tensor transpose_459_perm_0 = const()[name = string("transpose_459_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2295 = const()[name = string("concat_2295"), val = tensor([1, 64, 104])]; tensor transpose_459_cast_fp16 = transpose(perm = transpose_459_perm_0, x = var_4440_cast_fp16_5)[name = string("transpose_3535")]; tensor reshape_688_cast_fp16 = reshape(shape = concat_2295, x = transpose_459_cast_fp16)[name = string("reshape_688_cast_fp16")]; bool matmul_229_transpose_x_0 = const()[name = string("matmul_229_transpose_x_0"), val = bool(false)]; bool matmul_229_transpose_y_0 = const()[name = string("matmul_229_transpose_y_0"), val = bool(false)]; tensor matmul_229_cast_fp16 = matmul(transpose_x = matmul_229_transpose_x_0, transpose_y = matmul_229_transpose_y_0, x = reshape_687_cast_fp16, y = reshape_688_cast_fp16)[name = string("matmul_229_cast_fp16")]; tensor concat_2299 = const()[name = string("concat_2299"), val = tensor([1, 1, 104, 104])]; tensor reshape_689_cast_fp16 = reshape(shape = concat_2299, x = matmul_229_cast_fp16)[name = string("reshape_689_cast_fp16")]; tensor transpose_2917_perm_0 = const()[name = string("transpose_2917_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2917 = transpose(perm = transpose_2917_perm_0, x = reshape_689_cast_fp16)[name = string("transpose_3534")]; tensor w_919_cast_fp16 = add(x = transpose_2917, y = transpose_2305)[name = string("w_919_cast_fp16")]; tensor var_4519_cast_fp16 = softmax(axis = var_4367, x = w_919_cast_fp16)[name = string("op_4519_cast_fp16")]; string var_4521_equation_0 = const()[name = string("op_4521_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4521_cast_fp16 = einsum(equation = var_4521_equation_0, values = (var_4457_cast_fp16_5, var_4519_cast_fp16))[name = string("op_4521_cast_fp16")]; tensor transpose_460_perm_0 = const()[name = string("transpose_460_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2304 = const()[name = string("concat_2304"), val = tensor([1, 104, 64])]; tensor transpose_460_cast_fp16 = transpose(perm = transpose_460_perm_0, x = var_4423_cast_fp16_6)[name = string("transpose_3533")]; tensor reshape_690_cast_fp16 = reshape(shape = concat_2304, x = transpose_460_cast_fp16)[name = string("reshape_690_cast_fp16")]; tensor transpose_461_perm_0 = const()[name = string("transpose_461_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2305 = const()[name = string("concat_2305"), val = tensor([1, 64, 104])]; tensor transpose_461_cast_fp16 = transpose(perm = transpose_461_perm_0, x = var_4440_cast_fp16_6)[name = string("transpose_3532")]; tensor reshape_691_cast_fp16 = reshape(shape = concat_2305, x = transpose_461_cast_fp16)[name = string("reshape_691_cast_fp16")]; bool matmul_230_transpose_x_0 = const()[name = string("matmul_230_transpose_x_0"), val = bool(false)]; bool matmul_230_transpose_y_0 = const()[name = string("matmul_230_transpose_y_0"), val = bool(false)]; tensor matmul_230_cast_fp16 = matmul(transpose_x = matmul_230_transpose_x_0, transpose_y = matmul_230_transpose_y_0, x = reshape_690_cast_fp16, y = reshape_691_cast_fp16)[name = string("matmul_230_cast_fp16")]; tensor concat_2309 = const()[name = string("concat_2309"), val = tensor([1, 1, 104, 104])]; tensor reshape_692_cast_fp16 = reshape(shape = concat_2309, x = matmul_230_cast_fp16)[name = string("reshape_692_cast_fp16")]; tensor transpose_2918_perm_0 = const()[name = string("transpose_2918_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2918 = transpose(perm = transpose_2918_perm_0, x = reshape_692_cast_fp16)[name = string("transpose_3531")]; tensor w_923_cast_fp16 = add(x = transpose_2918, y = transpose_2305)[name = string("w_923_cast_fp16")]; tensor var_4527_cast_fp16 = softmax(axis = var_4367, x = w_923_cast_fp16)[name = string("op_4527_cast_fp16")]; string var_4529_equation_0 = const()[name = string("op_4529_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4529_cast_fp16 = einsum(equation = var_4529_equation_0, values = (var_4457_cast_fp16_6, var_4527_cast_fp16))[name = string("op_4529_cast_fp16")]; tensor transpose_462_perm_0 = const()[name = string("transpose_462_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2314 = const()[name = string("concat_2314"), val = tensor([1, 104, 64])]; tensor transpose_462_cast_fp16 = transpose(perm = transpose_462_perm_0, x = var_4423_cast_fp16_7)[name = string("transpose_3530")]; tensor reshape_693_cast_fp16 = reshape(shape = concat_2314, x = transpose_462_cast_fp16)[name = string("reshape_693_cast_fp16")]; tensor transpose_463_perm_0 = const()[name = string("transpose_463_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2315 = const()[name = string("concat_2315"), val = tensor([1, 64, 104])]; tensor transpose_463_cast_fp16 = transpose(perm = transpose_463_perm_0, x = var_4440_cast_fp16_7)[name = string("transpose_3529")]; tensor reshape_694_cast_fp16 = reshape(shape = concat_2315, x = transpose_463_cast_fp16)[name = string("reshape_694_cast_fp16")]; bool matmul_231_transpose_x_0 = const()[name = string("matmul_231_transpose_x_0"), val = bool(false)]; bool matmul_231_transpose_y_0 = const()[name = string("matmul_231_transpose_y_0"), val = bool(false)]; tensor matmul_231_cast_fp16 = matmul(transpose_x = matmul_231_transpose_x_0, transpose_y = matmul_231_transpose_y_0, x = reshape_693_cast_fp16, y = reshape_694_cast_fp16)[name = string("matmul_231_cast_fp16")]; tensor concat_2319 = const()[name = string("concat_2319"), val = tensor([1, 1, 104, 104])]; tensor reshape_695_cast_fp16 = reshape(shape = concat_2319, x = matmul_231_cast_fp16)[name = string("reshape_695_cast_fp16")]; tensor transpose_2919_perm_0 = const()[name = string("transpose_2919_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2919 = transpose(perm = transpose_2919_perm_0, x = reshape_695_cast_fp16)[name = string("transpose_3528")]; tensor w_927_cast_fp16 = add(x = transpose_2919, y = transpose_2305)[name = string("w_927_cast_fp16")]; tensor var_4535_cast_fp16 = softmax(axis = var_4367, x = w_927_cast_fp16)[name = string("op_4535_cast_fp16")]; string var_4537_equation_0 = const()[name = string("op_4537_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4537_cast_fp16 = einsum(equation = var_4537_equation_0, values = (var_4457_cast_fp16_7, var_4535_cast_fp16))[name = string("op_4537_cast_fp16")]; tensor transpose_464_perm_0 = const()[name = string("transpose_464_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2324 = const()[name = string("concat_2324"), val = tensor([1, 104, 64])]; tensor transpose_464_cast_fp16 = transpose(perm = transpose_464_perm_0, x = var_4423_cast_fp16_8)[name = string("transpose_3527")]; tensor reshape_696_cast_fp16 = reshape(shape = concat_2324, x = transpose_464_cast_fp16)[name = string("reshape_696_cast_fp16")]; tensor transpose_465_perm_0 = const()[name = string("transpose_465_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2325 = const()[name = string("concat_2325"), val = tensor([1, 64, 104])]; tensor transpose_465_cast_fp16 = transpose(perm = transpose_465_perm_0, x = var_4440_cast_fp16_8)[name = string("transpose_3526")]; tensor reshape_697_cast_fp16 = reshape(shape = concat_2325, x = transpose_465_cast_fp16)[name = string("reshape_697_cast_fp16")]; bool matmul_232_transpose_x_0 = const()[name = string("matmul_232_transpose_x_0"), val = bool(false)]; bool matmul_232_transpose_y_0 = const()[name = string("matmul_232_transpose_y_0"), val = bool(false)]; tensor matmul_232_cast_fp16 = matmul(transpose_x = matmul_232_transpose_x_0, transpose_y = matmul_232_transpose_y_0, x = reshape_696_cast_fp16, y = reshape_697_cast_fp16)[name = string("matmul_232_cast_fp16")]; tensor concat_2329 = const()[name = string("concat_2329"), val = tensor([1, 1, 104, 104])]; tensor reshape_698_cast_fp16 = reshape(shape = concat_2329, x = matmul_232_cast_fp16)[name = string("reshape_698_cast_fp16")]; tensor transpose_2920_perm_0 = const()[name = string("transpose_2920_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2920 = transpose(perm = transpose_2920_perm_0, x = reshape_698_cast_fp16)[name = string("transpose_3525")]; tensor w_931_cast_fp16 = add(x = transpose_2920, y = transpose_2305)[name = string("w_931_cast_fp16")]; tensor var_4543_cast_fp16 = softmax(axis = var_4367, x = w_931_cast_fp16)[name = string("op_4543_cast_fp16")]; string var_4545_equation_0 = const()[name = string("op_4545_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4545_cast_fp16 = einsum(equation = var_4545_equation_0, values = (var_4457_cast_fp16_8, var_4543_cast_fp16))[name = string("op_4545_cast_fp16")]; tensor transpose_466_perm_0 = const()[name = string("transpose_466_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2334 = const()[name = string("concat_2334"), val = tensor([1, 104, 64])]; tensor transpose_466_cast_fp16 = transpose(perm = transpose_466_perm_0, x = var_4423_cast_fp16_9)[name = string("transpose_3524")]; tensor reshape_699_cast_fp16 = reshape(shape = concat_2334, x = transpose_466_cast_fp16)[name = string("reshape_699_cast_fp16")]; tensor transpose_467_perm_0 = const()[name = string("transpose_467_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2335 = const()[name = string("concat_2335"), val = tensor([1, 64, 104])]; tensor transpose_467_cast_fp16 = transpose(perm = transpose_467_perm_0, x = var_4440_cast_fp16_9)[name = string("transpose_3523")]; tensor reshape_700_cast_fp16 = reshape(shape = concat_2335, x = transpose_467_cast_fp16)[name = string("reshape_700_cast_fp16")]; bool matmul_233_transpose_x_0 = const()[name = string("matmul_233_transpose_x_0"), val = bool(false)]; bool matmul_233_transpose_y_0 = const()[name = string("matmul_233_transpose_y_0"), val = bool(false)]; tensor matmul_233_cast_fp16 = matmul(transpose_x = matmul_233_transpose_x_0, transpose_y = matmul_233_transpose_y_0, x = reshape_699_cast_fp16, y = reshape_700_cast_fp16)[name = string("matmul_233_cast_fp16")]; tensor concat_2339 = const()[name = string("concat_2339"), val = tensor([1, 1, 104, 104])]; tensor reshape_701_cast_fp16 = reshape(shape = concat_2339, x = matmul_233_cast_fp16)[name = string("reshape_701_cast_fp16")]; tensor transpose_2921_perm_0 = const()[name = string("transpose_2921_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2921 = transpose(perm = transpose_2921_perm_0, x = reshape_701_cast_fp16)[name = string("transpose_3522")]; tensor w_935_cast_fp16 = add(x = transpose_2921, y = transpose_2305)[name = string("w_935_cast_fp16")]; tensor var_4551_cast_fp16 = softmax(axis = var_4367, x = w_935_cast_fp16)[name = string("op_4551_cast_fp16")]; string var_4553_equation_0 = const()[name = string("op_4553_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4553_cast_fp16 = einsum(equation = var_4553_equation_0, values = (var_4457_cast_fp16_9, var_4551_cast_fp16))[name = string("op_4553_cast_fp16")]; tensor transpose_468_perm_0 = const()[name = string("transpose_468_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2344 = const()[name = string("concat_2344"), val = tensor([1, 104, 64])]; tensor transpose_468_cast_fp16 = transpose(perm = transpose_468_perm_0, x = var_4423_cast_fp16_10)[name = string("transpose_3521")]; tensor reshape_702_cast_fp16 = reshape(shape = concat_2344, x = transpose_468_cast_fp16)[name = string("reshape_702_cast_fp16")]; tensor transpose_469_perm_0 = const()[name = string("transpose_469_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2345 = const()[name = string("concat_2345"), val = tensor([1, 64, 104])]; tensor transpose_469_cast_fp16 = transpose(perm = transpose_469_perm_0, x = var_4440_cast_fp16_10)[name = string("transpose_3520")]; tensor reshape_703_cast_fp16 = reshape(shape = concat_2345, x = transpose_469_cast_fp16)[name = string("reshape_703_cast_fp16")]; bool matmul_234_transpose_x_0 = const()[name = string("matmul_234_transpose_x_0"), val = bool(false)]; bool matmul_234_transpose_y_0 = const()[name = string("matmul_234_transpose_y_0"), val = bool(false)]; tensor matmul_234_cast_fp16 = matmul(transpose_x = matmul_234_transpose_x_0, transpose_y = matmul_234_transpose_y_0, x = reshape_702_cast_fp16, y = reshape_703_cast_fp16)[name = string("matmul_234_cast_fp16")]; tensor concat_2349 = const()[name = string("concat_2349"), val = tensor([1, 1, 104, 104])]; tensor reshape_704_cast_fp16 = reshape(shape = concat_2349, x = matmul_234_cast_fp16)[name = string("reshape_704_cast_fp16")]; tensor transpose_2922_perm_0 = const()[name = string("transpose_2922_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2922 = transpose(perm = transpose_2922_perm_0, x = reshape_704_cast_fp16)[name = string("transpose_3519")]; tensor w_939_cast_fp16 = add(x = transpose_2922, y = transpose_2305)[name = string("w_939_cast_fp16")]; tensor var_4559_cast_fp16 = softmax(axis = var_4367, x = w_939_cast_fp16)[name = string("op_4559_cast_fp16")]; string var_4561_equation_0 = const()[name = string("op_4561_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4561_cast_fp16 = einsum(equation = var_4561_equation_0, values = (var_4457_cast_fp16_10, var_4559_cast_fp16))[name = string("op_4561_cast_fp16")]; tensor transpose_470_perm_0 = const()[name = string("transpose_470_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2354 = const()[name = string("concat_2354"), val = tensor([1, 104, 64])]; tensor transpose_470_cast_fp16 = transpose(perm = transpose_470_perm_0, x = var_4423_cast_fp16_11)[name = string("transpose_3518")]; tensor reshape_705_cast_fp16 = reshape(shape = concat_2354, x = transpose_470_cast_fp16)[name = string("reshape_705_cast_fp16")]; tensor transpose_471_perm_0 = const()[name = string("transpose_471_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2355 = const()[name = string("concat_2355"), val = tensor([1, 64, 104])]; tensor transpose_471_cast_fp16 = transpose(perm = transpose_471_perm_0, x = var_4440_cast_fp16_11)[name = string("transpose_3517")]; tensor reshape_706_cast_fp16 = reshape(shape = concat_2355, x = transpose_471_cast_fp16)[name = string("reshape_706_cast_fp16")]; bool matmul_235_transpose_x_0 = const()[name = string("matmul_235_transpose_x_0"), val = bool(false)]; bool matmul_235_transpose_y_0 = const()[name = string("matmul_235_transpose_y_0"), val = bool(false)]; tensor matmul_235_cast_fp16 = matmul(transpose_x = matmul_235_transpose_x_0, transpose_y = matmul_235_transpose_y_0, x = reshape_705_cast_fp16, y = reshape_706_cast_fp16)[name = string("matmul_235_cast_fp16")]; tensor concat_2359 = const()[name = string("concat_2359"), val = tensor([1, 1, 104, 104])]; tensor reshape_707_cast_fp16 = reshape(shape = concat_2359, x = matmul_235_cast_fp16)[name = string("reshape_707_cast_fp16")]; tensor transpose_2923_perm_0 = const()[name = string("transpose_2923_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2923 = transpose(perm = transpose_2923_perm_0, x = reshape_707_cast_fp16)[name = string("transpose_3516")]; tensor w_943_cast_fp16 = add(x = transpose_2923, y = transpose_2305)[name = string("w_943_cast_fp16")]; tensor var_4567_cast_fp16 = softmax(axis = var_4367, x = w_943_cast_fp16)[name = string("op_4567_cast_fp16")]; string var_4569_equation_0 = const()[name = string("op_4569_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4569_cast_fp16 = einsum(equation = var_4569_equation_0, values = (var_4457_cast_fp16_11, var_4567_cast_fp16))[name = string("op_4569_cast_fp16")]; tensor transpose_472_perm_0 = const()[name = string("transpose_472_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2364 = const()[name = string("concat_2364"), val = tensor([1, 104, 64])]; tensor transpose_472_cast_fp16 = transpose(perm = transpose_472_perm_0, x = var_4423_cast_fp16_12)[name = string("transpose_3515")]; tensor reshape_708_cast_fp16 = reshape(shape = concat_2364, x = transpose_472_cast_fp16)[name = string("reshape_708_cast_fp16")]; tensor transpose_473_perm_0 = const()[name = string("transpose_473_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2365 = const()[name = string("concat_2365"), val = tensor([1, 64, 104])]; tensor transpose_473_cast_fp16 = transpose(perm = transpose_473_perm_0, x = var_4440_cast_fp16_12)[name = string("transpose_3514")]; tensor reshape_709_cast_fp16 = reshape(shape = concat_2365, x = transpose_473_cast_fp16)[name = string("reshape_709_cast_fp16")]; bool matmul_236_transpose_x_0 = const()[name = string("matmul_236_transpose_x_0"), val = bool(false)]; bool matmul_236_transpose_y_0 = const()[name = string("matmul_236_transpose_y_0"), val = bool(false)]; tensor matmul_236_cast_fp16 = matmul(transpose_x = matmul_236_transpose_x_0, transpose_y = matmul_236_transpose_y_0, x = reshape_708_cast_fp16, y = reshape_709_cast_fp16)[name = string("matmul_236_cast_fp16")]; tensor concat_2369 = const()[name = string("concat_2369"), val = tensor([1, 1, 104, 104])]; tensor reshape_710_cast_fp16 = reshape(shape = concat_2369, x = matmul_236_cast_fp16)[name = string("reshape_710_cast_fp16")]; tensor transpose_2924_perm_0 = const()[name = string("transpose_2924_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2924 = transpose(perm = transpose_2924_perm_0, x = reshape_710_cast_fp16)[name = string("transpose_3513")]; tensor w_947_cast_fp16 = add(x = transpose_2924, y = transpose_2305)[name = string("w_947_cast_fp16")]; tensor var_4575_cast_fp16 = softmax(axis = var_4367, x = w_947_cast_fp16)[name = string("op_4575_cast_fp16")]; string var_4577_equation_0 = const()[name = string("op_4577_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4577_cast_fp16 = einsum(equation = var_4577_equation_0, values = (var_4457_cast_fp16_12, var_4575_cast_fp16))[name = string("op_4577_cast_fp16")]; tensor transpose_474_perm_0 = const()[name = string("transpose_474_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2374 = const()[name = string("concat_2374"), val = tensor([1, 104, 64])]; tensor transpose_474_cast_fp16 = transpose(perm = transpose_474_perm_0, x = var_4423_cast_fp16_13)[name = string("transpose_3512")]; tensor reshape_711_cast_fp16 = reshape(shape = concat_2374, x = transpose_474_cast_fp16)[name = string("reshape_711_cast_fp16")]; tensor transpose_475_perm_0 = const()[name = string("transpose_475_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2375 = const()[name = string("concat_2375"), val = tensor([1, 64, 104])]; tensor transpose_475_cast_fp16 = transpose(perm = transpose_475_perm_0, x = var_4440_cast_fp16_13)[name = string("transpose_3511")]; tensor reshape_712_cast_fp16 = reshape(shape = concat_2375, x = transpose_475_cast_fp16)[name = string("reshape_712_cast_fp16")]; bool matmul_237_transpose_x_0 = const()[name = string("matmul_237_transpose_x_0"), val = bool(false)]; bool matmul_237_transpose_y_0 = const()[name = string("matmul_237_transpose_y_0"), val = bool(false)]; tensor matmul_237_cast_fp16 = matmul(transpose_x = matmul_237_transpose_x_0, transpose_y = matmul_237_transpose_y_0, x = reshape_711_cast_fp16, y = reshape_712_cast_fp16)[name = string("matmul_237_cast_fp16")]; tensor concat_2379 = const()[name = string("concat_2379"), val = tensor([1, 1, 104, 104])]; tensor reshape_713_cast_fp16 = reshape(shape = concat_2379, x = matmul_237_cast_fp16)[name = string("reshape_713_cast_fp16")]; tensor transpose_2925_perm_0 = const()[name = string("transpose_2925_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2925 = transpose(perm = transpose_2925_perm_0, x = reshape_713_cast_fp16)[name = string("transpose_3510")]; tensor w_951_cast_fp16 = add(x = transpose_2925, y = transpose_2305)[name = string("w_951_cast_fp16")]; tensor var_4583_cast_fp16 = softmax(axis = var_4367, x = w_951_cast_fp16)[name = string("op_4583_cast_fp16")]; string var_4585_equation_0 = const()[name = string("op_4585_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4585_cast_fp16 = einsum(equation = var_4585_equation_0, values = (var_4457_cast_fp16_13, var_4583_cast_fp16))[name = string("op_4585_cast_fp16")]; tensor transpose_476_perm_0 = const()[name = string("transpose_476_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2384 = const()[name = string("concat_2384"), val = tensor([1, 104, 64])]; tensor transpose_476_cast_fp16 = transpose(perm = transpose_476_perm_0, x = var_4423_cast_fp16_14)[name = string("transpose_3509")]; tensor reshape_714_cast_fp16 = reshape(shape = concat_2384, x = transpose_476_cast_fp16)[name = string("reshape_714_cast_fp16")]; tensor transpose_477_perm_0 = const()[name = string("transpose_477_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2385 = const()[name = string("concat_2385"), val = tensor([1, 64, 104])]; tensor transpose_477_cast_fp16 = transpose(perm = transpose_477_perm_0, x = var_4440_cast_fp16_14)[name = string("transpose_3508")]; tensor reshape_715_cast_fp16 = reshape(shape = concat_2385, x = transpose_477_cast_fp16)[name = string("reshape_715_cast_fp16")]; bool matmul_238_transpose_x_0 = const()[name = string("matmul_238_transpose_x_0"), val = bool(false)]; bool matmul_238_transpose_y_0 = const()[name = string("matmul_238_transpose_y_0"), val = bool(false)]; tensor matmul_238_cast_fp16 = matmul(transpose_x = matmul_238_transpose_x_0, transpose_y = matmul_238_transpose_y_0, x = reshape_714_cast_fp16, y = reshape_715_cast_fp16)[name = string("matmul_238_cast_fp16")]; tensor concat_2389 = const()[name = string("concat_2389"), val = tensor([1, 1, 104, 104])]; tensor reshape_716_cast_fp16 = reshape(shape = concat_2389, x = matmul_238_cast_fp16)[name = string("reshape_716_cast_fp16")]; tensor transpose_2926_perm_0 = const()[name = string("transpose_2926_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2926 = transpose(perm = transpose_2926_perm_0, x = reshape_716_cast_fp16)[name = string("transpose_3507")]; tensor w_955_cast_fp16 = add(x = transpose_2926, y = transpose_2305)[name = string("w_955_cast_fp16")]; tensor var_4591_cast_fp16 = softmax(axis = var_4367, x = w_955_cast_fp16)[name = string("op_4591_cast_fp16")]; string var_4593_equation_0 = const()[name = string("op_4593_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4593_cast_fp16 = einsum(equation = var_4593_equation_0, values = (var_4457_cast_fp16_14, var_4591_cast_fp16))[name = string("op_4593_cast_fp16")]; tensor transpose_478_perm_0 = const()[name = string("transpose_478_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2394 = const()[name = string("concat_2394"), val = tensor([1, 104, 64])]; tensor transpose_478_cast_fp16 = transpose(perm = transpose_478_perm_0, x = var_4423_cast_fp16_15)[name = string("transpose_3506")]; tensor reshape_717_cast_fp16 = reshape(shape = concat_2394, x = transpose_478_cast_fp16)[name = string("reshape_717_cast_fp16")]; tensor transpose_479_perm_0 = const()[name = string("transpose_479_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2395 = const()[name = string("concat_2395"), val = tensor([1, 64, 104])]; tensor transpose_479_cast_fp16 = transpose(perm = transpose_479_perm_0, x = var_4440_cast_fp16_15)[name = string("transpose_3505")]; tensor reshape_718_cast_fp16 = reshape(shape = concat_2395, x = transpose_479_cast_fp16)[name = string("reshape_718_cast_fp16")]; bool matmul_239_transpose_x_0 = const()[name = string("matmul_239_transpose_x_0"), val = bool(false)]; bool matmul_239_transpose_y_0 = const()[name = string("matmul_239_transpose_y_0"), val = bool(false)]; tensor matmul_239_cast_fp16 = matmul(transpose_x = matmul_239_transpose_x_0, transpose_y = matmul_239_transpose_y_0, x = reshape_717_cast_fp16, y = reshape_718_cast_fp16)[name = string("matmul_239_cast_fp16")]; tensor concat_2399 = const()[name = string("concat_2399"), val = tensor([1, 1, 104, 104])]; tensor reshape_719_cast_fp16 = reshape(shape = concat_2399, x = matmul_239_cast_fp16)[name = string("reshape_719_cast_fp16")]; tensor transpose_2927_perm_0 = const()[name = string("transpose_2927_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2927 = transpose(perm = transpose_2927_perm_0, x = reshape_719_cast_fp16)[name = string("transpose_3504")]; tensor w_959_cast_fp16 = add(x = transpose_2927, y = transpose_2305)[name = string("w_959_cast_fp16")]; tensor var_4599_cast_fp16 = softmax(axis = var_4367, x = w_959_cast_fp16)[name = string("op_4599_cast_fp16")]; string var_4601_equation_0 = const()[name = string("op_4601_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4601_cast_fp16 = einsum(equation = var_4601_equation_0, values = (var_4457_cast_fp16_15, var_4599_cast_fp16))[name = string("op_4601_cast_fp16")]; bool input_123_interleave_0 = const()[name = string("input_123_interleave_0"), val = bool(false)]; tensor input_123_cast_fp16 = concat(axis = var_4367, interleave = input_123_interleave_0, values = (var_4481_cast_fp16, var_4489_cast_fp16, var_4497_cast_fp16, var_4505_cast_fp16, var_4513_cast_fp16, var_4521_cast_fp16, var_4529_cast_fp16, var_4537_cast_fp16, var_4545_cast_fp16, var_4553_cast_fp16, var_4561_cast_fp16, var_4569_cast_fp16, var_4577_cast_fp16, var_4585_cast_fp16, var_4593_cast_fp16, var_4601_cast_fp16))[name = string("input_123_cast_fp16")]; string var_4610_pad_type_0 = const()[name = string("op_4610_pad_type_0"), val = string("valid")]; tensor var_4610_strides_0 = const()[name = string("op_4610_strides_0"), val = tensor([1, 1])]; tensor var_4610_pad_0 = const()[name = string("op_4610_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4610_dilations_0 = const()[name = string("op_4610_dilations_0"), val = tensor([1, 1])]; int32 var_4610_groups_0 = const()[name = string("op_4610_groups_0"), val = int32(1)]; tensor layers_14_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_14_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(383082560)))]; tensor layers_14_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_14_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(385179776)))]; tensor var_4610_cast_fp16 = conv(bias = layers_14_self_attn_out_proj_bias_to_fp16, dilations = var_4610_dilations_0, groups = var_4610_groups_0, pad = var_4610_pad_0, pad_type = var_4610_pad_type_0, strides = var_4610_strides_0, weight = layers_14_self_attn_out_proj_weight_to_fp16, x = input_123_cast_fp16)[name = string("op_4610_cast_fp16")]; tensor x_157_cast_fp16 = add(x = x_153_cast_fp16, y = var_4610_cast_fp16)[name = string("x_157_cast_fp16")]; tensor mu_59_axes_0 = const()[name = string("mu_59_axes_0"), val = tensor([1])]; bool mu_59_keep_dims_0 = const()[name = string("mu_59_keep_dims_0"), val = bool(true)]; tensor mu_59_cast_fp16 = reduce_mean(axes = mu_59_axes_0, keep_dims = mu_59_keep_dims_0, x = x_157_cast_fp16)[name = string("mu_59_cast_fp16")]; tensor var_4616_cast_fp16 = sub(x = x_157_cast_fp16, y = mu_59_cast_fp16)[name = string("op_4616_cast_fp16")]; fp16 var_4370_promoted_1_to_fp16 = const()[name = string("op_4370_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_4617_cast_fp16 = pow(x = var_4616_cast_fp16, y = var_4370_promoted_1_to_fp16)[name = string("op_4617_cast_fp16")]; tensor var_59_axes_0 = const()[name = string("var_59_axes_0"), val = tensor([1])]; bool var_59_keep_dims_0 = const()[name = string("var_59_keep_dims_0"), val = bool(true)]; tensor var_59_cast_fp16 = reduce_mean(axes = var_59_axes_0, keep_dims = var_59_keep_dims_0, x = var_4617_cast_fp16)[name = string("var_59_cast_fp16")]; fp16 var_4621_to_fp16 = const()[name = string("op_4621_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4622_cast_fp16 = add(x = var_59_cast_fp16, y = var_4621_to_fp16)[name = string("op_4622_cast_fp16")]; fp32 var_4623_epsilon_0 = const()[name = string("op_4623_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4623_cast_fp16 = rsqrt(epsilon = var_4623_epsilon_0, x = var_4622_cast_fp16)[name = string("op_4623_cast_fp16")]; tensor x_159_cast_fp16 = mul(x = var_4616_cast_fp16, y = var_4623_cast_fp16)[name = string("x_159_cast_fp16")]; tensor input_125_gamma_0_to_fp16 = const()[name = string("input_125_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(385181888)))]; tensor input_125_beta_0_to_fp16 = const()[name = string("input_125_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(385184000)))]; fp16 input_125_epsilon_0_to_fp16 = const()[name = string("input_125_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_125_cast_fp16 = batch_norm(beta = input_125_beta_0_to_fp16, epsilon = input_125_epsilon_0_to_fp16, gamma = input_125_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_159_cast_fp16)[name = string("input_125_cast_fp16")]; string x_161_pad_type_0 = const()[name = string("x_161_pad_type_0"), val = string("valid")]; tensor x_161_strides_0 = const()[name = string("x_161_strides_0"), val = tensor([1, 1])]; tensor x_161_pad_0 = const()[name = string("x_161_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_161_dilations_0 = const()[name = string("x_161_dilations_0"), val = tensor([1, 1])]; int32 x_161_groups_0 = const()[name = string("x_161_groups_0"), val = int32(1)]; tensor layers_14_fc1_weight_to_fp16 = const()[name = string("layers_14_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(385186112)))]; tensor layers_14_fc1_bias_to_fp16 = const()[name = string("layers_14_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(393574784)))]; tensor x_161_cast_fp16 = conv(bias = layers_14_fc1_bias_to_fp16, dilations = x_161_dilations_0, groups = x_161_groups_0, pad = x_161_pad_0, pad_type = x_161_pad_type_0, strides = x_161_strides_0, weight = layers_14_fc1_weight_to_fp16, x = input_125_cast_fp16)[name = string("x_161_cast_fp16")]; fp16 var_4638_to_fp16 = const()[name = string("op_4638_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_4639_cast_fp16 = mul(x = x_161_cast_fp16, y = var_4638_to_fp16)[name = string("op_4639_cast_fp16")]; tensor var_4640_cast_fp16 = mul(x = var_4639_cast_fp16, y = x_161_cast_fp16)[name = string("op_4640_cast_fp16")]; tensor var_4641_cast_fp16 = mul(x = var_4640_cast_fp16, y = x_161_cast_fp16)[name = string("op_4641_cast_fp16")]; tensor var_4642_cast_fp16 = add(x = x_161_cast_fp16, y = var_4641_cast_fp16)[name = string("op_4642_cast_fp16")]; fp16 var_4643_to_fp16 = const()[name = string("op_4643_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_35_cast_fp16 = mul(x = var_4642_cast_fp16, y = var_4643_to_fp16)[name = string("u_35_cast_fp16")]; fp16 var_4645_to_fp16 = const()[name = string("op_4645_to_fp16"), val = fp16(0x1p-1)]; tensor var_4646_cast_fp16 = mul(x = x_161_cast_fp16, y = var_4645_to_fp16)[name = string("op_4646_cast_fp16")]; tensor var_4647_cast_fp16 = tanh(x = u_35_cast_fp16)[name = string("op_4647_cast_fp16")]; fp16 var_4648_to_fp16 = const()[name = string("op_4648_to_fp16"), val = fp16(0x1p+0)]; tensor var_4649_cast_fp16 = add(x = var_4647_cast_fp16, y = var_4648_to_fp16)[name = string("op_4649_cast_fp16")]; tensor input_127_cast_fp16 = mul(x = var_4646_cast_fp16, y = var_4649_cast_fp16)[name = string("input_127_cast_fp16")]; string h_29_pad_type_0 = const()[name = string("h_29_pad_type_0"), val = string("valid")]; tensor h_29_strides_0 = const()[name = string("h_29_strides_0"), val = tensor([1, 1])]; tensor h_29_pad_0 = const()[name = string("h_29_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_29_dilations_0 = const()[name = string("h_29_dilations_0"), val = tensor([1, 1])]; int32 h_29_groups_0 = const()[name = string("h_29_groups_0"), val = int32(1)]; tensor layers_14_fc2_weight_to_fp16 = const()[name = string("layers_14_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(393583040)))]; tensor layers_14_fc2_bias_to_fp16 = const()[name = string("layers_14_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(401971712)))]; tensor h_29_cast_fp16 = conv(bias = layers_14_fc2_bias_to_fp16, dilations = h_29_dilations_0, groups = h_29_groups_0, pad = h_29_pad_0, pad_type = h_29_pad_type_0, strides = h_29_strides_0, weight = layers_14_fc2_weight_to_fp16, x = input_127_cast_fp16)[name = string("h_29_cast_fp16")]; tensor x_163_cast_fp16 = add(x = x_157_cast_fp16, y = h_29_cast_fp16)[name = string("x_163_cast_fp16")]; int32 var_4665 = const()[name = string("op_4665"), val = int32(1)]; tensor mu_61_axes_0 = const()[name = string("mu_61_axes_0"), val = tensor([1])]; bool mu_61_keep_dims_0 = const()[name = string("mu_61_keep_dims_0"), val = bool(true)]; tensor mu_61_cast_fp16 = reduce_mean(axes = mu_61_axes_0, keep_dims = mu_61_keep_dims_0, x = x_163_cast_fp16)[name = string("mu_61_cast_fp16")]; tensor var_4679_cast_fp16 = sub(x = x_163_cast_fp16, y = mu_61_cast_fp16)[name = string("op_4679_cast_fp16")]; fp16 var_4668_promoted_to_fp16 = const()[name = string("op_4668_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_4680_cast_fp16 = pow(x = var_4679_cast_fp16, y = var_4668_promoted_to_fp16)[name = string("op_4680_cast_fp16")]; tensor var_61_axes_0 = const()[name = string("var_61_axes_0"), val = tensor([1])]; bool var_61_keep_dims_0 = const()[name = string("var_61_keep_dims_0"), val = bool(true)]; tensor var_61_cast_fp16 = reduce_mean(axes = var_61_axes_0, keep_dims = var_61_keep_dims_0, x = var_4680_cast_fp16)[name = string("var_61_cast_fp16")]; fp16 var_4684_to_fp16 = const()[name = string("op_4684_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4685_cast_fp16 = add(x = var_61_cast_fp16, y = var_4684_to_fp16)[name = string("op_4685_cast_fp16")]; fp32 var_4686_epsilon_0 = const()[name = string("op_4686_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4686_cast_fp16 = rsqrt(epsilon = var_4686_epsilon_0, x = var_4685_cast_fp16)[name = string("op_4686_cast_fp16")]; tensor x_165_cast_fp16 = mul(x = var_4679_cast_fp16, y = var_4686_cast_fp16)[name = string("x_165_cast_fp16")]; tensor input_129_gamma_0_to_fp16 = const()[name = string("input_129_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(401973824)))]; tensor input_129_beta_0_to_fp16 = const()[name = string("input_129_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(401975936)))]; fp16 input_129_epsilon_0_to_fp16 = const()[name = string("input_129_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_129_cast_fp16 = batch_norm(beta = input_129_beta_0_to_fp16, epsilon = input_129_epsilon_0_to_fp16, gamma = input_129_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_165_cast_fp16)[name = string("input_129_cast_fp16")]; string var_4704_pad_type_0 = const()[name = string("op_4704_pad_type_0"), val = string("valid")]; tensor var_4704_strides_0 = const()[name = string("op_4704_strides_0"), val = tensor([1, 1])]; tensor var_4704_pad_0 = const()[name = string("op_4704_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4704_dilations_0 = const()[name = string("op_4704_dilations_0"), val = tensor([1, 1])]; int32 var_4704_groups_0 = const()[name = string("op_4704_groups_0"), val = int32(1)]; tensor var_4706_weight_0_to_fp16 = const()[name = string("op_4706_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(401978048)))]; tensor var_4706_bias_0_to_fp16 = const()[name = string("op_4706_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(404075264)))]; tensor var_4706_cast_fp16 = conv(bias = var_4706_bias_0_to_fp16, dilations = var_4704_dilations_0, groups = var_4704_groups_0, pad = var_4704_pad_0, pad_type = var_4704_pad_type_0, strides = var_4704_strides_0, weight = var_4706_weight_0_to_fp16, x = input_129_cast_fp16)[name = string("op_4706_cast_fp16")]; string var_4713_pad_type_0 = const()[name = string("op_4713_pad_type_0"), val = string("valid")]; tensor var_4713_strides_0 = const()[name = string("op_4713_strides_0"), val = tensor([1, 1])]; tensor var_4713_pad_0 = const()[name = string("op_4713_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4713_dilations_0 = const()[name = string("op_4713_dilations_0"), val = tensor([1, 1])]; int32 var_4713_groups_0 = const()[name = string("op_4713_groups_0"), val = int32(1)]; tensor layers_15_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_15_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(404077376)))]; tensor layers_15_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_15_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(406174592)))]; tensor var_4713_cast_fp16 = conv(bias = layers_15_self_attn_k_proj_bias_to_fp16, dilations = var_4713_dilations_0, groups = var_4713_groups_0, pad = var_4713_pad_0, pad_type = var_4713_pad_type_0, strides = var_4713_strides_0, weight = layers_15_self_attn_k_proj_weight_to_fp16, x = input_129_cast_fp16)[name = string("op_4713_cast_fp16")]; string var_4720_pad_type_0 = const()[name = string("op_4720_pad_type_0"), val = string("valid")]; tensor var_4720_strides_0 = const()[name = string("op_4720_strides_0"), val = tensor([1, 1])]; tensor var_4720_pad_0 = const()[name = string("op_4720_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4720_dilations_0 = const()[name = string("op_4720_dilations_0"), val = tensor([1, 1])]; int32 var_4720_groups_0 = const()[name = string("op_4720_groups_0"), val = int32(1)]; tensor layers_15_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_15_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(406176704)))]; tensor layers_15_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_15_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(408273920)))]; tensor var_4720_cast_fp16 = conv(bias = layers_15_self_attn_v_proj_bias_to_fp16, dilations = var_4720_dilations_0, groups = var_4720_groups_0, pad = var_4720_pad_0, pad_type = var_4720_pad_type_0, strides = var_4720_strides_0, weight = layers_15_self_attn_v_proj_weight_to_fp16, x = input_129_cast_fp16)[name = string("op_4720_cast_fp16")]; tensor tile_45 = const()[name = string("tile_45"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(408276032)))]; int32 var_4721_axis_0 = const()[name = string("op_4721_axis_0"), val = int32(1)]; tensor var_4721_cast_fp16_0, tensor var_4721_cast_fp16_1, tensor var_4721_cast_fp16_2, tensor var_4721_cast_fp16_3, tensor var_4721_cast_fp16_4, tensor var_4721_cast_fp16_5, tensor var_4721_cast_fp16_6, tensor var_4721_cast_fp16_7, tensor var_4721_cast_fp16_8, tensor var_4721_cast_fp16_9, tensor var_4721_cast_fp16_10, tensor var_4721_cast_fp16_11, tensor var_4721_cast_fp16_12, tensor var_4721_cast_fp16_13, tensor var_4721_cast_fp16_14, tensor var_4721_cast_fp16_15 = split(axis = var_4721_axis_0, split_sizes = tile_45, x = var_4706_cast_fp16)[name = string("op_4721_cast_fp16")]; tensor tile_46 = const()[name = string("tile_46"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(408276160)))]; int32 var_4738_axis_0 = const()[name = string("op_4738_axis_0"), val = int32(1)]; tensor var_4738_cast_fp16_0, tensor var_4738_cast_fp16_1, tensor var_4738_cast_fp16_2, tensor var_4738_cast_fp16_3, tensor var_4738_cast_fp16_4, tensor var_4738_cast_fp16_5, tensor var_4738_cast_fp16_6, tensor var_4738_cast_fp16_7, tensor var_4738_cast_fp16_8, tensor var_4738_cast_fp16_9, tensor var_4738_cast_fp16_10, tensor var_4738_cast_fp16_11, tensor var_4738_cast_fp16_12, tensor var_4738_cast_fp16_13, tensor var_4738_cast_fp16_14, tensor var_4738_cast_fp16_15 = split(axis = var_4738_axis_0, split_sizes = tile_46, x = var_4713_cast_fp16)[name = string("op_4738_cast_fp16")]; tensor tile_47 = const()[name = string("tile_47"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(408276288)))]; int32 var_4755_axis_0 = const()[name = string("op_4755_axis_0"), val = int32(1)]; tensor var_4755_cast_fp16_0, tensor var_4755_cast_fp16_1, tensor var_4755_cast_fp16_2, tensor var_4755_cast_fp16_3, tensor var_4755_cast_fp16_4, tensor var_4755_cast_fp16_5, tensor var_4755_cast_fp16_6, tensor var_4755_cast_fp16_7, tensor var_4755_cast_fp16_8, tensor var_4755_cast_fp16_9, tensor var_4755_cast_fp16_10, tensor var_4755_cast_fp16_11, tensor var_4755_cast_fp16_12, tensor var_4755_cast_fp16_13, tensor var_4755_cast_fp16_14, tensor var_4755_cast_fp16_15 = split(axis = var_4755_axis_0, split_sizes = tile_47, x = var_4720_cast_fp16)[name = string("op_4755_cast_fp16")]; tensor transpose_480_perm_0 = const()[name = string("transpose_480_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2404 = const()[name = string("concat_2404"), val = tensor([1, 104, 64])]; tensor transpose_480_cast_fp16 = transpose(perm = transpose_480_perm_0, x = var_4721_cast_fp16_0)[name = string("transpose_3503")]; tensor reshape_720_cast_fp16 = reshape(shape = concat_2404, x = transpose_480_cast_fp16)[name = string("reshape_720_cast_fp16")]; tensor transpose_481_perm_0 = const()[name = string("transpose_481_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2405 = const()[name = string("concat_2405"), val = tensor([1, 64, 104])]; tensor transpose_481_cast_fp16 = transpose(perm = transpose_481_perm_0, x = var_4738_cast_fp16_0)[name = string("transpose_3502")]; tensor reshape_721_cast_fp16 = reshape(shape = concat_2405, x = transpose_481_cast_fp16)[name = string("reshape_721_cast_fp16")]; bool matmul_240_transpose_x_0 = const()[name = string("matmul_240_transpose_x_0"), val = bool(false)]; bool matmul_240_transpose_y_0 = const()[name = string("matmul_240_transpose_y_0"), val = bool(false)]; tensor matmul_240_cast_fp16 = matmul(transpose_x = matmul_240_transpose_x_0, transpose_y = matmul_240_transpose_y_0, x = reshape_720_cast_fp16, y = reshape_721_cast_fp16)[name = string("matmul_240_cast_fp16")]; tensor concat_2409 = const()[name = string("concat_2409"), val = tensor([1, 1, 104, 104])]; tensor reshape_722_cast_fp16 = reshape(shape = concat_2409, x = matmul_240_cast_fp16)[name = string("reshape_722_cast_fp16")]; tensor transpose_2928_perm_0 = const()[name = string("transpose_2928_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2928 = transpose(perm = transpose_2928_perm_0, x = reshape_722_cast_fp16)[name = string("transpose_3501")]; tensor w_963_cast_fp16 = add(x = transpose_2928, y = transpose_2305)[name = string("w_963_cast_fp16")]; tensor var_4777_cast_fp16 = softmax(axis = var_4665, x = w_963_cast_fp16)[name = string("op_4777_cast_fp16")]; string var_4779_equation_0 = const()[name = string("op_4779_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4779_cast_fp16 = einsum(equation = var_4779_equation_0, values = (var_4755_cast_fp16_0, var_4777_cast_fp16))[name = string("op_4779_cast_fp16")]; tensor transpose_482_perm_0 = const()[name = string("transpose_482_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2414 = const()[name = string("concat_2414"), val = tensor([1, 104, 64])]; tensor transpose_482_cast_fp16 = transpose(perm = transpose_482_perm_0, x = var_4721_cast_fp16_1)[name = string("transpose_3500")]; tensor reshape_723_cast_fp16 = reshape(shape = concat_2414, x = transpose_482_cast_fp16)[name = string("reshape_723_cast_fp16")]; tensor transpose_483_perm_0 = const()[name = string("transpose_483_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2415 = const()[name = string("concat_2415"), val = tensor([1, 64, 104])]; tensor transpose_483_cast_fp16 = transpose(perm = transpose_483_perm_0, x = var_4738_cast_fp16_1)[name = string("transpose_3499")]; tensor reshape_724_cast_fp16 = reshape(shape = concat_2415, x = transpose_483_cast_fp16)[name = string("reshape_724_cast_fp16")]; bool matmul_241_transpose_x_0 = const()[name = string("matmul_241_transpose_x_0"), val = bool(false)]; bool matmul_241_transpose_y_0 = const()[name = string("matmul_241_transpose_y_0"), val = bool(false)]; tensor matmul_241_cast_fp16 = matmul(transpose_x = matmul_241_transpose_x_0, transpose_y = matmul_241_transpose_y_0, x = reshape_723_cast_fp16, y = reshape_724_cast_fp16)[name = string("matmul_241_cast_fp16")]; tensor concat_2419 = const()[name = string("concat_2419"), val = tensor([1, 1, 104, 104])]; tensor reshape_725_cast_fp16 = reshape(shape = concat_2419, x = matmul_241_cast_fp16)[name = string("reshape_725_cast_fp16")]; tensor transpose_2929_perm_0 = const()[name = string("transpose_2929_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2929 = transpose(perm = transpose_2929_perm_0, x = reshape_725_cast_fp16)[name = string("transpose_3498")]; tensor w_967_cast_fp16 = add(x = transpose_2929, y = transpose_2305)[name = string("w_967_cast_fp16")]; tensor var_4785_cast_fp16 = softmax(axis = var_4665, x = w_967_cast_fp16)[name = string("op_4785_cast_fp16")]; string var_4787_equation_0 = const()[name = string("op_4787_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4787_cast_fp16 = einsum(equation = var_4787_equation_0, values = (var_4755_cast_fp16_1, var_4785_cast_fp16))[name = string("op_4787_cast_fp16")]; tensor transpose_484_perm_0 = const()[name = string("transpose_484_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2424 = const()[name = string("concat_2424"), val = tensor([1, 104, 64])]; tensor transpose_484_cast_fp16 = transpose(perm = transpose_484_perm_0, x = var_4721_cast_fp16_2)[name = string("transpose_3497")]; tensor reshape_726_cast_fp16 = reshape(shape = concat_2424, x = transpose_484_cast_fp16)[name = string("reshape_726_cast_fp16")]; tensor transpose_485_perm_0 = const()[name = string("transpose_485_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2425 = const()[name = string("concat_2425"), val = tensor([1, 64, 104])]; tensor transpose_485_cast_fp16 = transpose(perm = transpose_485_perm_0, x = var_4738_cast_fp16_2)[name = string("transpose_3496")]; tensor reshape_727_cast_fp16 = reshape(shape = concat_2425, x = transpose_485_cast_fp16)[name = string("reshape_727_cast_fp16")]; bool matmul_242_transpose_x_0 = const()[name = string("matmul_242_transpose_x_0"), val = bool(false)]; bool matmul_242_transpose_y_0 = const()[name = string("matmul_242_transpose_y_0"), val = bool(false)]; tensor matmul_242_cast_fp16 = matmul(transpose_x = matmul_242_transpose_x_0, transpose_y = matmul_242_transpose_y_0, x = reshape_726_cast_fp16, y = reshape_727_cast_fp16)[name = string("matmul_242_cast_fp16")]; tensor concat_2429 = const()[name = string("concat_2429"), val = tensor([1, 1, 104, 104])]; tensor reshape_728_cast_fp16 = reshape(shape = concat_2429, x = matmul_242_cast_fp16)[name = string("reshape_728_cast_fp16")]; tensor transpose_2930_perm_0 = const()[name = string("transpose_2930_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2930 = transpose(perm = transpose_2930_perm_0, x = reshape_728_cast_fp16)[name = string("transpose_3495")]; tensor w_971_cast_fp16 = add(x = transpose_2930, y = transpose_2305)[name = string("w_971_cast_fp16")]; tensor var_4793_cast_fp16 = softmax(axis = var_4665, x = w_971_cast_fp16)[name = string("op_4793_cast_fp16")]; string var_4795_equation_0 = const()[name = string("op_4795_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4795_cast_fp16 = einsum(equation = var_4795_equation_0, values = (var_4755_cast_fp16_2, var_4793_cast_fp16))[name = string("op_4795_cast_fp16")]; tensor transpose_486_perm_0 = const()[name = string("transpose_486_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2434 = const()[name = string("concat_2434"), val = tensor([1, 104, 64])]; tensor transpose_486_cast_fp16 = transpose(perm = transpose_486_perm_0, x = var_4721_cast_fp16_3)[name = string("transpose_3494")]; tensor reshape_729_cast_fp16 = reshape(shape = concat_2434, x = transpose_486_cast_fp16)[name = string("reshape_729_cast_fp16")]; tensor transpose_487_perm_0 = const()[name = string("transpose_487_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2435 = const()[name = string("concat_2435"), val = tensor([1, 64, 104])]; tensor transpose_487_cast_fp16 = transpose(perm = transpose_487_perm_0, x = var_4738_cast_fp16_3)[name = string("transpose_3493")]; tensor reshape_730_cast_fp16 = reshape(shape = concat_2435, x = transpose_487_cast_fp16)[name = string("reshape_730_cast_fp16")]; bool matmul_243_transpose_x_0 = const()[name = string("matmul_243_transpose_x_0"), val = bool(false)]; bool matmul_243_transpose_y_0 = const()[name = string("matmul_243_transpose_y_0"), val = bool(false)]; tensor matmul_243_cast_fp16 = matmul(transpose_x = matmul_243_transpose_x_0, transpose_y = matmul_243_transpose_y_0, x = reshape_729_cast_fp16, y = reshape_730_cast_fp16)[name = string("matmul_243_cast_fp16")]; tensor concat_2439 = const()[name = string("concat_2439"), val = tensor([1, 1, 104, 104])]; tensor reshape_731_cast_fp16 = reshape(shape = concat_2439, x = matmul_243_cast_fp16)[name = string("reshape_731_cast_fp16")]; tensor transpose_2931_perm_0 = const()[name = string("transpose_2931_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2931 = transpose(perm = transpose_2931_perm_0, x = reshape_731_cast_fp16)[name = string("transpose_3492")]; tensor w_975_cast_fp16 = add(x = transpose_2931, y = transpose_2305)[name = string("w_975_cast_fp16")]; tensor var_4801_cast_fp16 = softmax(axis = var_4665, x = w_975_cast_fp16)[name = string("op_4801_cast_fp16")]; string var_4803_equation_0 = const()[name = string("op_4803_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4803_cast_fp16 = einsum(equation = var_4803_equation_0, values = (var_4755_cast_fp16_3, var_4801_cast_fp16))[name = string("op_4803_cast_fp16")]; tensor transpose_488_perm_0 = const()[name = string("transpose_488_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2444 = const()[name = string("concat_2444"), val = tensor([1, 104, 64])]; tensor transpose_488_cast_fp16 = transpose(perm = transpose_488_perm_0, x = var_4721_cast_fp16_4)[name = string("transpose_3491")]; tensor reshape_732_cast_fp16 = reshape(shape = concat_2444, x = transpose_488_cast_fp16)[name = string("reshape_732_cast_fp16")]; tensor transpose_489_perm_0 = const()[name = string("transpose_489_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2445 = const()[name = string("concat_2445"), val = tensor([1, 64, 104])]; tensor transpose_489_cast_fp16 = transpose(perm = transpose_489_perm_0, x = var_4738_cast_fp16_4)[name = string("transpose_3490")]; tensor reshape_733_cast_fp16 = reshape(shape = concat_2445, x = transpose_489_cast_fp16)[name = string("reshape_733_cast_fp16")]; bool matmul_244_transpose_x_0 = const()[name = string("matmul_244_transpose_x_0"), val = bool(false)]; bool matmul_244_transpose_y_0 = const()[name = string("matmul_244_transpose_y_0"), val = bool(false)]; tensor matmul_244_cast_fp16 = matmul(transpose_x = matmul_244_transpose_x_0, transpose_y = matmul_244_transpose_y_0, x = reshape_732_cast_fp16, y = reshape_733_cast_fp16)[name = string("matmul_244_cast_fp16")]; tensor concat_2449 = const()[name = string("concat_2449"), val = tensor([1, 1, 104, 104])]; tensor reshape_734_cast_fp16 = reshape(shape = concat_2449, x = matmul_244_cast_fp16)[name = string("reshape_734_cast_fp16")]; tensor transpose_2932_perm_0 = const()[name = string("transpose_2932_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2932 = transpose(perm = transpose_2932_perm_0, x = reshape_734_cast_fp16)[name = string("transpose_3489")]; tensor w_979_cast_fp16 = add(x = transpose_2932, y = transpose_2305)[name = string("w_979_cast_fp16")]; tensor var_4809_cast_fp16 = softmax(axis = var_4665, x = w_979_cast_fp16)[name = string("op_4809_cast_fp16")]; string var_4811_equation_0 = const()[name = string("op_4811_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4811_cast_fp16 = einsum(equation = var_4811_equation_0, values = (var_4755_cast_fp16_4, var_4809_cast_fp16))[name = string("op_4811_cast_fp16")]; tensor transpose_490_perm_0 = const()[name = string("transpose_490_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2454 = const()[name = string("concat_2454"), val = tensor([1, 104, 64])]; tensor transpose_490_cast_fp16 = transpose(perm = transpose_490_perm_0, x = var_4721_cast_fp16_5)[name = string("transpose_3488")]; tensor reshape_735_cast_fp16 = reshape(shape = concat_2454, x = transpose_490_cast_fp16)[name = string("reshape_735_cast_fp16")]; tensor transpose_491_perm_0 = const()[name = string("transpose_491_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2455 = const()[name = string("concat_2455"), val = tensor([1, 64, 104])]; tensor transpose_491_cast_fp16 = transpose(perm = transpose_491_perm_0, x = var_4738_cast_fp16_5)[name = string("transpose_3487")]; tensor reshape_736_cast_fp16 = reshape(shape = concat_2455, x = transpose_491_cast_fp16)[name = string("reshape_736_cast_fp16")]; bool matmul_245_transpose_x_0 = const()[name = string("matmul_245_transpose_x_0"), val = bool(false)]; bool matmul_245_transpose_y_0 = const()[name = string("matmul_245_transpose_y_0"), val = bool(false)]; tensor matmul_245_cast_fp16 = matmul(transpose_x = matmul_245_transpose_x_0, transpose_y = matmul_245_transpose_y_0, x = reshape_735_cast_fp16, y = reshape_736_cast_fp16)[name = string("matmul_245_cast_fp16")]; tensor concat_2459 = const()[name = string("concat_2459"), val = tensor([1, 1, 104, 104])]; tensor reshape_737_cast_fp16 = reshape(shape = concat_2459, x = matmul_245_cast_fp16)[name = string("reshape_737_cast_fp16")]; tensor transpose_2933_perm_0 = const()[name = string("transpose_2933_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2933 = transpose(perm = transpose_2933_perm_0, x = reshape_737_cast_fp16)[name = string("transpose_3486")]; tensor w_983_cast_fp16 = add(x = transpose_2933, y = transpose_2305)[name = string("w_983_cast_fp16")]; tensor var_4817_cast_fp16 = softmax(axis = var_4665, x = w_983_cast_fp16)[name = string("op_4817_cast_fp16")]; string var_4819_equation_0 = const()[name = string("op_4819_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4819_cast_fp16 = einsum(equation = var_4819_equation_0, values = (var_4755_cast_fp16_5, var_4817_cast_fp16))[name = string("op_4819_cast_fp16")]; tensor transpose_492_perm_0 = const()[name = string("transpose_492_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2464 = const()[name = string("concat_2464"), val = tensor([1, 104, 64])]; tensor transpose_492_cast_fp16 = transpose(perm = transpose_492_perm_0, x = var_4721_cast_fp16_6)[name = string("transpose_3485")]; tensor reshape_738_cast_fp16 = reshape(shape = concat_2464, x = transpose_492_cast_fp16)[name = string("reshape_738_cast_fp16")]; tensor transpose_493_perm_0 = const()[name = string("transpose_493_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2465 = const()[name = string("concat_2465"), val = tensor([1, 64, 104])]; tensor transpose_493_cast_fp16 = transpose(perm = transpose_493_perm_0, x = var_4738_cast_fp16_6)[name = string("transpose_3484")]; tensor reshape_739_cast_fp16 = reshape(shape = concat_2465, x = transpose_493_cast_fp16)[name = string("reshape_739_cast_fp16")]; bool matmul_246_transpose_x_0 = const()[name = string("matmul_246_transpose_x_0"), val = bool(false)]; bool matmul_246_transpose_y_0 = const()[name = string("matmul_246_transpose_y_0"), val = bool(false)]; tensor matmul_246_cast_fp16 = matmul(transpose_x = matmul_246_transpose_x_0, transpose_y = matmul_246_transpose_y_0, x = reshape_738_cast_fp16, y = reshape_739_cast_fp16)[name = string("matmul_246_cast_fp16")]; tensor concat_2469 = const()[name = string("concat_2469"), val = tensor([1, 1, 104, 104])]; tensor reshape_740_cast_fp16 = reshape(shape = concat_2469, x = matmul_246_cast_fp16)[name = string("reshape_740_cast_fp16")]; tensor transpose_2934_perm_0 = const()[name = string("transpose_2934_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2934 = transpose(perm = transpose_2934_perm_0, x = reshape_740_cast_fp16)[name = string("transpose_3483")]; tensor w_987_cast_fp16 = add(x = transpose_2934, y = transpose_2305)[name = string("w_987_cast_fp16")]; tensor var_4825_cast_fp16 = softmax(axis = var_4665, x = w_987_cast_fp16)[name = string("op_4825_cast_fp16")]; string var_4827_equation_0 = const()[name = string("op_4827_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4827_cast_fp16 = einsum(equation = var_4827_equation_0, values = (var_4755_cast_fp16_6, var_4825_cast_fp16))[name = string("op_4827_cast_fp16")]; tensor transpose_494_perm_0 = const()[name = string("transpose_494_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2474 = const()[name = string("concat_2474"), val = tensor([1, 104, 64])]; tensor transpose_494_cast_fp16 = transpose(perm = transpose_494_perm_0, x = var_4721_cast_fp16_7)[name = string("transpose_3482")]; tensor reshape_741_cast_fp16 = reshape(shape = concat_2474, x = transpose_494_cast_fp16)[name = string("reshape_741_cast_fp16")]; tensor transpose_495_perm_0 = const()[name = string("transpose_495_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2475 = const()[name = string("concat_2475"), val = tensor([1, 64, 104])]; tensor transpose_495_cast_fp16 = transpose(perm = transpose_495_perm_0, x = var_4738_cast_fp16_7)[name = string("transpose_3481")]; tensor reshape_742_cast_fp16 = reshape(shape = concat_2475, x = transpose_495_cast_fp16)[name = string("reshape_742_cast_fp16")]; bool matmul_247_transpose_x_0 = const()[name = string("matmul_247_transpose_x_0"), val = bool(false)]; bool matmul_247_transpose_y_0 = const()[name = string("matmul_247_transpose_y_0"), val = bool(false)]; tensor matmul_247_cast_fp16 = matmul(transpose_x = matmul_247_transpose_x_0, transpose_y = matmul_247_transpose_y_0, x = reshape_741_cast_fp16, y = reshape_742_cast_fp16)[name = string("matmul_247_cast_fp16")]; tensor concat_2479 = const()[name = string("concat_2479"), val = tensor([1, 1, 104, 104])]; tensor reshape_743_cast_fp16 = reshape(shape = concat_2479, x = matmul_247_cast_fp16)[name = string("reshape_743_cast_fp16")]; tensor transpose_2935_perm_0 = const()[name = string("transpose_2935_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2935 = transpose(perm = transpose_2935_perm_0, x = reshape_743_cast_fp16)[name = string("transpose_3480")]; tensor w_991_cast_fp16 = add(x = transpose_2935, y = transpose_2305)[name = string("w_991_cast_fp16")]; tensor var_4833_cast_fp16 = softmax(axis = var_4665, x = w_991_cast_fp16)[name = string("op_4833_cast_fp16")]; string var_4835_equation_0 = const()[name = string("op_4835_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4835_cast_fp16 = einsum(equation = var_4835_equation_0, values = (var_4755_cast_fp16_7, var_4833_cast_fp16))[name = string("op_4835_cast_fp16")]; tensor transpose_496_perm_0 = const()[name = string("transpose_496_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2484 = const()[name = string("concat_2484"), val = tensor([1, 104, 64])]; tensor transpose_496_cast_fp16 = transpose(perm = transpose_496_perm_0, x = var_4721_cast_fp16_8)[name = string("transpose_3479")]; tensor reshape_744_cast_fp16 = reshape(shape = concat_2484, x = transpose_496_cast_fp16)[name = string("reshape_744_cast_fp16")]; tensor transpose_497_perm_0 = const()[name = string("transpose_497_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2485 = const()[name = string("concat_2485"), val = tensor([1, 64, 104])]; tensor transpose_497_cast_fp16 = transpose(perm = transpose_497_perm_0, x = var_4738_cast_fp16_8)[name = string("transpose_3478")]; tensor reshape_745_cast_fp16 = reshape(shape = concat_2485, x = transpose_497_cast_fp16)[name = string("reshape_745_cast_fp16")]; bool matmul_248_transpose_x_0 = const()[name = string("matmul_248_transpose_x_0"), val = bool(false)]; bool matmul_248_transpose_y_0 = const()[name = string("matmul_248_transpose_y_0"), val = bool(false)]; tensor matmul_248_cast_fp16 = matmul(transpose_x = matmul_248_transpose_x_0, transpose_y = matmul_248_transpose_y_0, x = reshape_744_cast_fp16, y = reshape_745_cast_fp16)[name = string("matmul_248_cast_fp16")]; tensor concat_2489 = const()[name = string("concat_2489"), val = tensor([1, 1, 104, 104])]; tensor reshape_746_cast_fp16 = reshape(shape = concat_2489, x = matmul_248_cast_fp16)[name = string("reshape_746_cast_fp16")]; tensor transpose_2936_perm_0 = const()[name = string("transpose_2936_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2936 = transpose(perm = transpose_2936_perm_0, x = reshape_746_cast_fp16)[name = string("transpose_3477")]; tensor w_995_cast_fp16 = add(x = transpose_2936, y = transpose_2305)[name = string("w_995_cast_fp16")]; tensor var_4841_cast_fp16 = softmax(axis = var_4665, x = w_995_cast_fp16)[name = string("op_4841_cast_fp16")]; string var_4843_equation_0 = const()[name = string("op_4843_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4843_cast_fp16 = einsum(equation = var_4843_equation_0, values = (var_4755_cast_fp16_8, var_4841_cast_fp16))[name = string("op_4843_cast_fp16")]; tensor transpose_498_perm_0 = const()[name = string("transpose_498_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2494 = const()[name = string("concat_2494"), val = tensor([1, 104, 64])]; tensor transpose_498_cast_fp16 = transpose(perm = transpose_498_perm_0, x = var_4721_cast_fp16_9)[name = string("transpose_3476")]; tensor reshape_747_cast_fp16 = reshape(shape = concat_2494, x = transpose_498_cast_fp16)[name = string("reshape_747_cast_fp16")]; tensor transpose_499_perm_0 = const()[name = string("transpose_499_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2495 = const()[name = string("concat_2495"), val = tensor([1, 64, 104])]; tensor transpose_499_cast_fp16 = transpose(perm = transpose_499_perm_0, x = var_4738_cast_fp16_9)[name = string("transpose_3475")]; tensor reshape_748_cast_fp16 = reshape(shape = concat_2495, x = transpose_499_cast_fp16)[name = string("reshape_748_cast_fp16")]; bool matmul_249_transpose_x_0 = const()[name = string("matmul_249_transpose_x_0"), val = bool(false)]; bool matmul_249_transpose_y_0 = const()[name = string("matmul_249_transpose_y_0"), val = bool(false)]; tensor matmul_249_cast_fp16 = matmul(transpose_x = matmul_249_transpose_x_0, transpose_y = matmul_249_transpose_y_0, x = reshape_747_cast_fp16, y = reshape_748_cast_fp16)[name = string("matmul_249_cast_fp16")]; tensor concat_2499 = const()[name = string("concat_2499"), val = tensor([1, 1, 104, 104])]; tensor reshape_749_cast_fp16 = reshape(shape = concat_2499, x = matmul_249_cast_fp16)[name = string("reshape_749_cast_fp16")]; tensor transpose_2937_perm_0 = const()[name = string("transpose_2937_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2937 = transpose(perm = transpose_2937_perm_0, x = reshape_749_cast_fp16)[name = string("transpose_3474")]; tensor w_999_cast_fp16 = add(x = transpose_2937, y = transpose_2305)[name = string("w_999_cast_fp16")]; tensor var_4849_cast_fp16 = softmax(axis = var_4665, x = w_999_cast_fp16)[name = string("op_4849_cast_fp16")]; string var_4851_equation_0 = const()[name = string("op_4851_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4851_cast_fp16 = einsum(equation = var_4851_equation_0, values = (var_4755_cast_fp16_9, var_4849_cast_fp16))[name = string("op_4851_cast_fp16")]; tensor transpose_500_perm_0 = const()[name = string("transpose_500_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2504 = const()[name = string("concat_2504"), val = tensor([1, 104, 64])]; tensor transpose_500_cast_fp16 = transpose(perm = transpose_500_perm_0, x = var_4721_cast_fp16_10)[name = string("transpose_3473")]; tensor reshape_750_cast_fp16 = reshape(shape = concat_2504, x = transpose_500_cast_fp16)[name = string("reshape_750_cast_fp16")]; tensor transpose_501_perm_0 = const()[name = string("transpose_501_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2505 = const()[name = string("concat_2505"), val = tensor([1, 64, 104])]; tensor transpose_501_cast_fp16 = transpose(perm = transpose_501_perm_0, x = var_4738_cast_fp16_10)[name = string("transpose_3472")]; tensor reshape_751_cast_fp16 = reshape(shape = concat_2505, x = transpose_501_cast_fp16)[name = string("reshape_751_cast_fp16")]; bool matmul_250_transpose_x_0 = const()[name = string("matmul_250_transpose_x_0"), val = bool(false)]; bool matmul_250_transpose_y_0 = const()[name = string("matmul_250_transpose_y_0"), val = bool(false)]; tensor matmul_250_cast_fp16 = matmul(transpose_x = matmul_250_transpose_x_0, transpose_y = matmul_250_transpose_y_0, x = reshape_750_cast_fp16, y = reshape_751_cast_fp16)[name = string("matmul_250_cast_fp16")]; tensor concat_2509 = const()[name = string("concat_2509"), val = tensor([1, 1, 104, 104])]; tensor reshape_752_cast_fp16 = reshape(shape = concat_2509, x = matmul_250_cast_fp16)[name = string("reshape_752_cast_fp16")]; tensor transpose_2938_perm_0 = const()[name = string("transpose_2938_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2938 = transpose(perm = transpose_2938_perm_0, x = reshape_752_cast_fp16)[name = string("transpose_3471")]; tensor w_1003_cast_fp16 = add(x = transpose_2938, y = transpose_2305)[name = string("w_1003_cast_fp16")]; tensor var_4857_cast_fp16 = softmax(axis = var_4665, x = w_1003_cast_fp16)[name = string("op_4857_cast_fp16")]; string var_4859_equation_0 = const()[name = string("op_4859_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4859_cast_fp16 = einsum(equation = var_4859_equation_0, values = (var_4755_cast_fp16_10, var_4857_cast_fp16))[name = string("op_4859_cast_fp16")]; tensor transpose_502_perm_0 = const()[name = string("transpose_502_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2514 = const()[name = string("concat_2514"), val = tensor([1, 104, 64])]; tensor transpose_502_cast_fp16 = transpose(perm = transpose_502_perm_0, x = var_4721_cast_fp16_11)[name = string("transpose_3470")]; tensor reshape_753_cast_fp16 = reshape(shape = concat_2514, x = transpose_502_cast_fp16)[name = string("reshape_753_cast_fp16")]; tensor transpose_503_perm_0 = const()[name = string("transpose_503_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2515 = const()[name = string("concat_2515"), val = tensor([1, 64, 104])]; tensor transpose_503_cast_fp16 = transpose(perm = transpose_503_perm_0, x = var_4738_cast_fp16_11)[name = string("transpose_3469")]; tensor reshape_754_cast_fp16 = reshape(shape = concat_2515, x = transpose_503_cast_fp16)[name = string("reshape_754_cast_fp16")]; bool matmul_251_transpose_x_0 = const()[name = string("matmul_251_transpose_x_0"), val = bool(false)]; bool matmul_251_transpose_y_0 = const()[name = string("matmul_251_transpose_y_0"), val = bool(false)]; tensor matmul_251_cast_fp16 = matmul(transpose_x = matmul_251_transpose_x_0, transpose_y = matmul_251_transpose_y_0, x = reshape_753_cast_fp16, y = reshape_754_cast_fp16)[name = string("matmul_251_cast_fp16")]; tensor concat_2519 = const()[name = string("concat_2519"), val = tensor([1, 1, 104, 104])]; tensor reshape_755_cast_fp16 = reshape(shape = concat_2519, x = matmul_251_cast_fp16)[name = string("reshape_755_cast_fp16")]; tensor transpose_2939_perm_0 = const()[name = string("transpose_2939_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2939 = transpose(perm = transpose_2939_perm_0, x = reshape_755_cast_fp16)[name = string("transpose_3468")]; tensor w_1007_cast_fp16 = add(x = transpose_2939, y = transpose_2305)[name = string("w_1007_cast_fp16")]; tensor var_4865_cast_fp16 = softmax(axis = var_4665, x = w_1007_cast_fp16)[name = string("op_4865_cast_fp16")]; string var_4867_equation_0 = const()[name = string("op_4867_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4867_cast_fp16 = einsum(equation = var_4867_equation_0, values = (var_4755_cast_fp16_11, var_4865_cast_fp16))[name = string("op_4867_cast_fp16")]; tensor transpose_504_perm_0 = const()[name = string("transpose_504_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2524 = const()[name = string("concat_2524"), val = tensor([1, 104, 64])]; tensor transpose_504_cast_fp16 = transpose(perm = transpose_504_perm_0, x = var_4721_cast_fp16_12)[name = string("transpose_3467")]; tensor reshape_756_cast_fp16 = reshape(shape = concat_2524, x = transpose_504_cast_fp16)[name = string("reshape_756_cast_fp16")]; tensor transpose_505_perm_0 = const()[name = string("transpose_505_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2525 = const()[name = string("concat_2525"), val = tensor([1, 64, 104])]; tensor transpose_505_cast_fp16 = transpose(perm = transpose_505_perm_0, x = var_4738_cast_fp16_12)[name = string("transpose_3466")]; tensor reshape_757_cast_fp16 = reshape(shape = concat_2525, x = transpose_505_cast_fp16)[name = string("reshape_757_cast_fp16")]; bool matmul_252_transpose_x_0 = const()[name = string("matmul_252_transpose_x_0"), val = bool(false)]; bool matmul_252_transpose_y_0 = const()[name = string("matmul_252_transpose_y_0"), val = bool(false)]; tensor matmul_252_cast_fp16 = matmul(transpose_x = matmul_252_transpose_x_0, transpose_y = matmul_252_transpose_y_0, x = reshape_756_cast_fp16, y = reshape_757_cast_fp16)[name = string("matmul_252_cast_fp16")]; tensor concat_2529 = const()[name = string("concat_2529"), val = tensor([1, 1, 104, 104])]; tensor reshape_758_cast_fp16 = reshape(shape = concat_2529, x = matmul_252_cast_fp16)[name = string("reshape_758_cast_fp16")]; tensor transpose_2940_perm_0 = const()[name = string("transpose_2940_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2940 = transpose(perm = transpose_2940_perm_0, x = reshape_758_cast_fp16)[name = string("transpose_3465")]; tensor w_1011_cast_fp16 = add(x = transpose_2940, y = transpose_2305)[name = string("w_1011_cast_fp16")]; tensor var_4873_cast_fp16 = softmax(axis = var_4665, x = w_1011_cast_fp16)[name = string("op_4873_cast_fp16")]; string var_4875_equation_0 = const()[name = string("op_4875_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4875_cast_fp16 = einsum(equation = var_4875_equation_0, values = (var_4755_cast_fp16_12, var_4873_cast_fp16))[name = string("op_4875_cast_fp16")]; tensor transpose_506_perm_0 = const()[name = string("transpose_506_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2534 = const()[name = string("concat_2534"), val = tensor([1, 104, 64])]; tensor transpose_506_cast_fp16 = transpose(perm = transpose_506_perm_0, x = var_4721_cast_fp16_13)[name = string("transpose_3464")]; tensor reshape_759_cast_fp16 = reshape(shape = concat_2534, x = transpose_506_cast_fp16)[name = string("reshape_759_cast_fp16")]; tensor transpose_507_perm_0 = const()[name = string("transpose_507_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2535 = const()[name = string("concat_2535"), val = tensor([1, 64, 104])]; tensor transpose_507_cast_fp16 = transpose(perm = transpose_507_perm_0, x = var_4738_cast_fp16_13)[name = string("transpose_3463")]; tensor reshape_760_cast_fp16 = reshape(shape = concat_2535, x = transpose_507_cast_fp16)[name = string("reshape_760_cast_fp16")]; bool matmul_253_transpose_x_0 = const()[name = string("matmul_253_transpose_x_0"), val = bool(false)]; bool matmul_253_transpose_y_0 = const()[name = string("matmul_253_transpose_y_0"), val = bool(false)]; tensor matmul_253_cast_fp16 = matmul(transpose_x = matmul_253_transpose_x_0, transpose_y = matmul_253_transpose_y_0, x = reshape_759_cast_fp16, y = reshape_760_cast_fp16)[name = string("matmul_253_cast_fp16")]; tensor concat_2539 = const()[name = string("concat_2539"), val = tensor([1, 1, 104, 104])]; tensor reshape_761_cast_fp16 = reshape(shape = concat_2539, x = matmul_253_cast_fp16)[name = string("reshape_761_cast_fp16")]; tensor transpose_2941_perm_0 = const()[name = string("transpose_2941_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2941 = transpose(perm = transpose_2941_perm_0, x = reshape_761_cast_fp16)[name = string("transpose_3462")]; tensor w_1015_cast_fp16 = add(x = transpose_2941, y = transpose_2305)[name = string("w_1015_cast_fp16")]; tensor var_4881_cast_fp16 = softmax(axis = var_4665, x = w_1015_cast_fp16)[name = string("op_4881_cast_fp16")]; string var_4883_equation_0 = const()[name = string("op_4883_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4883_cast_fp16 = einsum(equation = var_4883_equation_0, values = (var_4755_cast_fp16_13, var_4881_cast_fp16))[name = string("op_4883_cast_fp16")]; tensor transpose_508_perm_0 = const()[name = string("transpose_508_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2544 = const()[name = string("concat_2544"), val = tensor([1, 104, 64])]; tensor transpose_508_cast_fp16 = transpose(perm = transpose_508_perm_0, x = var_4721_cast_fp16_14)[name = string("transpose_3461")]; tensor reshape_762_cast_fp16 = reshape(shape = concat_2544, x = transpose_508_cast_fp16)[name = string("reshape_762_cast_fp16")]; tensor transpose_509_perm_0 = const()[name = string("transpose_509_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2545 = const()[name = string("concat_2545"), val = tensor([1, 64, 104])]; tensor transpose_509_cast_fp16 = transpose(perm = transpose_509_perm_0, x = var_4738_cast_fp16_14)[name = string("transpose_3460")]; tensor reshape_763_cast_fp16 = reshape(shape = concat_2545, x = transpose_509_cast_fp16)[name = string("reshape_763_cast_fp16")]; bool matmul_254_transpose_x_0 = const()[name = string("matmul_254_transpose_x_0"), val = bool(false)]; bool matmul_254_transpose_y_0 = const()[name = string("matmul_254_transpose_y_0"), val = bool(false)]; tensor matmul_254_cast_fp16 = matmul(transpose_x = matmul_254_transpose_x_0, transpose_y = matmul_254_transpose_y_0, x = reshape_762_cast_fp16, y = reshape_763_cast_fp16)[name = string("matmul_254_cast_fp16")]; tensor concat_2549 = const()[name = string("concat_2549"), val = tensor([1, 1, 104, 104])]; tensor reshape_764_cast_fp16 = reshape(shape = concat_2549, x = matmul_254_cast_fp16)[name = string("reshape_764_cast_fp16")]; tensor transpose_2942_perm_0 = const()[name = string("transpose_2942_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2942 = transpose(perm = transpose_2942_perm_0, x = reshape_764_cast_fp16)[name = string("transpose_3459")]; tensor w_1019_cast_fp16 = add(x = transpose_2942, y = transpose_2305)[name = string("w_1019_cast_fp16")]; tensor var_4889_cast_fp16 = softmax(axis = var_4665, x = w_1019_cast_fp16)[name = string("op_4889_cast_fp16")]; string var_4891_equation_0 = const()[name = string("op_4891_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4891_cast_fp16 = einsum(equation = var_4891_equation_0, values = (var_4755_cast_fp16_14, var_4889_cast_fp16))[name = string("op_4891_cast_fp16")]; tensor transpose_510_perm_0 = const()[name = string("transpose_510_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2554 = const()[name = string("concat_2554"), val = tensor([1, 104, 64])]; tensor transpose_510_cast_fp16 = transpose(perm = transpose_510_perm_0, x = var_4721_cast_fp16_15)[name = string("transpose_3458")]; tensor reshape_765_cast_fp16 = reshape(shape = concat_2554, x = transpose_510_cast_fp16)[name = string("reshape_765_cast_fp16")]; tensor transpose_511_perm_0 = const()[name = string("transpose_511_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2555 = const()[name = string("concat_2555"), val = tensor([1, 64, 104])]; tensor transpose_511_cast_fp16 = transpose(perm = transpose_511_perm_0, x = var_4738_cast_fp16_15)[name = string("transpose_3457")]; tensor reshape_766_cast_fp16 = reshape(shape = concat_2555, x = transpose_511_cast_fp16)[name = string("reshape_766_cast_fp16")]; bool matmul_255_transpose_x_0 = const()[name = string("matmul_255_transpose_x_0"), val = bool(false)]; bool matmul_255_transpose_y_0 = const()[name = string("matmul_255_transpose_y_0"), val = bool(false)]; tensor matmul_255_cast_fp16 = matmul(transpose_x = matmul_255_transpose_x_0, transpose_y = matmul_255_transpose_y_0, x = reshape_765_cast_fp16, y = reshape_766_cast_fp16)[name = string("matmul_255_cast_fp16")]; tensor concat_2559 = const()[name = string("concat_2559"), val = tensor([1, 1, 104, 104])]; tensor reshape_767_cast_fp16 = reshape(shape = concat_2559, x = matmul_255_cast_fp16)[name = string("reshape_767_cast_fp16")]; tensor transpose_2943_perm_0 = const()[name = string("transpose_2943_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2943 = transpose(perm = transpose_2943_perm_0, x = reshape_767_cast_fp16)[name = string("transpose_3456")]; tensor w_1023_cast_fp16 = add(x = transpose_2943, y = transpose_2305)[name = string("w_1023_cast_fp16")]; tensor var_4897_cast_fp16 = softmax(axis = var_4665, x = w_1023_cast_fp16)[name = string("op_4897_cast_fp16")]; string var_4899_equation_0 = const()[name = string("op_4899_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_4899_cast_fp16 = einsum(equation = var_4899_equation_0, values = (var_4755_cast_fp16_15, var_4897_cast_fp16))[name = string("op_4899_cast_fp16")]; bool input_131_interleave_0 = const()[name = string("input_131_interleave_0"), val = bool(false)]; tensor input_131_cast_fp16 = concat(axis = var_4665, interleave = input_131_interleave_0, values = (var_4779_cast_fp16, var_4787_cast_fp16, var_4795_cast_fp16, var_4803_cast_fp16, var_4811_cast_fp16, var_4819_cast_fp16, var_4827_cast_fp16, var_4835_cast_fp16, var_4843_cast_fp16, var_4851_cast_fp16, var_4859_cast_fp16, var_4867_cast_fp16, var_4875_cast_fp16, var_4883_cast_fp16, var_4891_cast_fp16, var_4899_cast_fp16))[name = string("input_131_cast_fp16")]; string var_4908_pad_type_0 = const()[name = string("op_4908_pad_type_0"), val = string("valid")]; tensor var_4908_strides_0 = const()[name = string("op_4908_strides_0"), val = tensor([1, 1])]; tensor var_4908_pad_0 = const()[name = string("op_4908_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_4908_dilations_0 = const()[name = string("op_4908_dilations_0"), val = tensor([1, 1])]; int32 var_4908_groups_0 = const()[name = string("op_4908_groups_0"), val = int32(1)]; tensor layers_15_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_15_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(408276416)))]; tensor layers_15_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_15_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(410373632)))]; tensor var_4908_cast_fp16 = conv(bias = layers_15_self_attn_out_proj_bias_to_fp16, dilations = var_4908_dilations_0, groups = var_4908_groups_0, pad = var_4908_pad_0, pad_type = var_4908_pad_type_0, strides = var_4908_strides_0, weight = layers_15_self_attn_out_proj_weight_to_fp16, x = input_131_cast_fp16)[name = string("op_4908_cast_fp16")]; tensor x_167_cast_fp16 = add(x = x_163_cast_fp16, y = var_4908_cast_fp16)[name = string("x_167_cast_fp16")]; tensor mu_63_axes_0 = const()[name = string("mu_63_axes_0"), val = tensor([1])]; bool mu_63_keep_dims_0 = const()[name = string("mu_63_keep_dims_0"), val = bool(true)]; tensor mu_63_cast_fp16 = reduce_mean(axes = mu_63_axes_0, keep_dims = mu_63_keep_dims_0, x = x_167_cast_fp16)[name = string("mu_63_cast_fp16")]; tensor var_4914_cast_fp16 = sub(x = x_167_cast_fp16, y = mu_63_cast_fp16)[name = string("op_4914_cast_fp16")]; fp16 var_4668_promoted_1_to_fp16 = const()[name = string("op_4668_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_4915_cast_fp16 = pow(x = var_4914_cast_fp16, y = var_4668_promoted_1_to_fp16)[name = string("op_4915_cast_fp16")]; tensor var_63_axes_0 = const()[name = string("var_63_axes_0"), val = tensor([1])]; bool var_63_keep_dims_0 = const()[name = string("var_63_keep_dims_0"), val = bool(true)]; tensor var_63_cast_fp16 = reduce_mean(axes = var_63_axes_0, keep_dims = var_63_keep_dims_0, x = var_4915_cast_fp16)[name = string("var_63_cast_fp16")]; fp16 var_4919_to_fp16 = const()[name = string("op_4919_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4920_cast_fp16 = add(x = var_63_cast_fp16, y = var_4919_to_fp16)[name = string("op_4920_cast_fp16")]; fp32 var_4921_epsilon_0 = const()[name = string("op_4921_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4921_cast_fp16 = rsqrt(epsilon = var_4921_epsilon_0, x = var_4920_cast_fp16)[name = string("op_4921_cast_fp16")]; tensor x_169_cast_fp16 = mul(x = var_4914_cast_fp16, y = var_4921_cast_fp16)[name = string("x_169_cast_fp16")]; tensor input_133_gamma_0_to_fp16 = const()[name = string("input_133_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(410375744)))]; tensor input_133_beta_0_to_fp16 = const()[name = string("input_133_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(410377856)))]; fp16 input_133_epsilon_0_to_fp16 = const()[name = string("input_133_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_133_cast_fp16 = batch_norm(beta = input_133_beta_0_to_fp16, epsilon = input_133_epsilon_0_to_fp16, gamma = input_133_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_169_cast_fp16)[name = string("input_133_cast_fp16")]; string x_171_pad_type_0 = const()[name = string("x_171_pad_type_0"), val = string("valid")]; tensor x_171_strides_0 = const()[name = string("x_171_strides_0"), val = tensor([1, 1])]; tensor x_171_pad_0 = const()[name = string("x_171_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_171_dilations_0 = const()[name = string("x_171_dilations_0"), val = tensor([1, 1])]; int32 x_171_groups_0 = const()[name = string("x_171_groups_0"), val = int32(1)]; tensor layers_15_fc1_weight_to_fp16 = const()[name = string("layers_15_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(410379968)))]; tensor layers_15_fc1_bias_to_fp16 = const()[name = string("layers_15_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(418768640)))]; tensor x_171_cast_fp16 = conv(bias = layers_15_fc1_bias_to_fp16, dilations = x_171_dilations_0, groups = x_171_groups_0, pad = x_171_pad_0, pad_type = x_171_pad_type_0, strides = x_171_strides_0, weight = layers_15_fc1_weight_to_fp16, x = input_133_cast_fp16)[name = string("x_171_cast_fp16")]; fp16 var_4936_to_fp16 = const()[name = string("op_4936_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_4937_cast_fp16 = mul(x = x_171_cast_fp16, y = var_4936_to_fp16)[name = string("op_4937_cast_fp16")]; tensor var_4938_cast_fp16 = mul(x = var_4937_cast_fp16, y = x_171_cast_fp16)[name = string("op_4938_cast_fp16")]; tensor var_4939_cast_fp16 = mul(x = var_4938_cast_fp16, y = x_171_cast_fp16)[name = string("op_4939_cast_fp16")]; tensor var_4940_cast_fp16 = add(x = x_171_cast_fp16, y = var_4939_cast_fp16)[name = string("op_4940_cast_fp16")]; fp16 var_4941_to_fp16 = const()[name = string("op_4941_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_37_cast_fp16 = mul(x = var_4940_cast_fp16, y = var_4941_to_fp16)[name = string("u_37_cast_fp16")]; fp16 var_4943_to_fp16 = const()[name = string("op_4943_to_fp16"), val = fp16(0x1p-1)]; tensor var_4944_cast_fp16 = mul(x = x_171_cast_fp16, y = var_4943_to_fp16)[name = string("op_4944_cast_fp16")]; tensor var_4945_cast_fp16 = tanh(x = u_37_cast_fp16)[name = string("op_4945_cast_fp16")]; fp16 var_4946_to_fp16 = const()[name = string("op_4946_to_fp16"), val = fp16(0x1p+0)]; tensor var_4947_cast_fp16 = add(x = var_4945_cast_fp16, y = var_4946_to_fp16)[name = string("op_4947_cast_fp16")]; tensor input_135_cast_fp16 = mul(x = var_4944_cast_fp16, y = var_4947_cast_fp16)[name = string("input_135_cast_fp16")]; string h_31_pad_type_0 = const()[name = string("h_31_pad_type_0"), val = string("valid")]; tensor h_31_strides_0 = const()[name = string("h_31_strides_0"), val = tensor([1, 1])]; tensor h_31_pad_0 = const()[name = string("h_31_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_31_dilations_0 = const()[name = string("h_31_dilations_0"), val = tensor([1, 1])]; int32 h_31_groups_0 = const()[name = string("h_31_groups_0"), val = int32(1)]; tensor layers_15_fc2_weight_to_fp16 = const()[name = string("layers_15_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(418776896)))]; tensor layers_15_fc2_bias_to_fp16 = const()[name = string("layers_15_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(427165568)))]; tensor h_31_cast_fp16 = conv(bias = layers_15_fc2_bias_to_fp16, dilations = h_31_dilations_0, groups = h_31_groups_0, pad = h_31_pad_0, pad_type = h_31_pad_type_0, strides = h_31_strides_0, weight = layers_15_fc2_weight_to_fp16, x = input_135_cast_fp16)[name = string("h_31_cast_fp16")]; tensor x_173_cast_fp16 = add(x = x_167_cast_fp16, y = h_31_cast_fp16)[name = string("x_173_cast_fp16")]; int32 var_4963 = const()[name = string("op_4963"), val = int32(1)]; tensor mu_65_axes_0 = const()[name = string("mu_65_axes_0"), val = tensor([1])]; bool mu_65_keep_dims_0 = const()[name = string("mu_65_keep_dims_0"), val = bool(true)]; tensor mu_65_cast_fp16 = reduce_mean(axes = mu_65_axes_0, keep_dims = mu_65_keep_dims_0, x = x_173_cast_fp16)[name = string("mu_65_cast_fp16")]; tensor var_4977_cast_fp16 = sub(x = x_173_cast_fp16, y = mu_65_cast_fp16)[name = string("op_4977_cast_fp16")]; fp16 var_4966_promoted_to_fp16 = const()[name = string("op_4966_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_4978_cast_fp16 = pow(x = var_4977_cast_fp16, y = var_4966_promoted_to_fp16)[name = string("op_4978_cast_fp16")]; tensor var_65_axes_0 = const()[name = string("var_65_axes_0"), val = tensor([1])]; bool var_65_keep_dims_0 = const()[name = string("var_65_keep_dims_0"), val = bool(true)]; tensor var_65_cast_fp16 = reduce_mean(axes = var_65_axes_0, keep_dims = var_65_keep_dims_0, x = var_4978_cast_fp16)[name = string("var_65_cast_fp16")]; fp16 var_4982_to_fp16 = const()[name = string("op_4982_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_4983_cast_fp16 = add(x = var_65_cast_fp16, y = var_4982_to_fp16)[name = string("op_4983_cast_fp16")]; fp32 var_4984_epsilon_0 = const()[name = string("op_4984_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_4984_cast_fp16 = rsqrt(epsilon = var_4984_epsilon_0, x = var_4983_cast_fp16)[name = string("op_4984_cast_fp16")]; tensor x_175_cast_fp16 = mul(x = var_4977_cast_fp16, y = var_4984_cast_fp16)[name = string("x_175_cast_fp16")]; tensor input_137_gamma_0_to_fp16 = const()[name = string("input_137_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(427167680)))]; tensor input_137_beta_0_to_fp16 = const()[name = string("input_137_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(427169792)))]; fp16 input_137_epsilon_0_to_fp16 = const()[name = string("input_137_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_137_cast_fp16 = batch_norm(beta = input_137_beta_0_to_fp16, epsilon = input_137_epsilon_0_to_fp16, gamma = input_137_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_175_cast_fp16)[name = string("input_137_cast_fp16")]; string var_5002_pad_type_0 = const()[name = string("op_5002_pad_type_0"), val = string("valid")]; tensor var_5002_strides_0 = const()[name = string("op_5002_strides_0"), val = tensor([1, 1])]; tensor var_5002_pad_0 = const()[name = string("op_5002_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5002_dilations_0 = const()[name = string("op_5002_dilations_0"), val = tensor([1, 1])]; int32 var_5002_groups_0 = const()[name = string("op_5002_groups_0"), val = int32(1)]; tensor var_5004_weight_0_to_fp16 = const()[name = string("op_5004_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(427171904)))]; tensor var_5004_bias_0_to_fp16 = const()[name = string("op_5004_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(429269120)))]; tensor var_5004_cast_fp16 = conv(bias = var_5004_bias_0_to_fp16, dilations = var_5002_dilations_0, groups = var_5002_groups_0, pad = var_5002_pad_0, pad_type = var_5002_pad_type_0, strides = var_5002_strides_0, weight = var_5004_weight_0_to_fp16, x = input_137_cast_fp16)[name = string("op_5004_cast_fp16")]; string var_5011_pad_type_0 = const()[name = string("op_5011_pad_type_0"), val = string("valid")]; tensor var_5011_strides_0 = const()[name = string("op_5011_strides_0"), val = tensor([1, 1])]; tensor var_5011_pad_0 = const()[name = string("op_5011_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5011_dilations_0 = const()[name = string("op_5011_dilations_0"), val = tensor([1, 1])]; int32 var_5011_groups_0 = const()[name = string("op_5011_groups_0"), val = int32(1)]; tensor layers_16_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_16_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(429271232)))]; tensor layers_16_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_16_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(431368448)))]; tensor var_5011_cast_fp16 = conv(bias = layers_16_self_attn_k_proj_bias_to_fp16, dilations = var_5011_dilations_0, groups = var_5011_groups_0, pad = var_5011_pad_0, pad_type = var_5011_pad_type_0, strides = var_5011_strides_0, weight = layers_16_self_attn_k_proj_weight_to_fp16, x = input_137_cast_fp16)[name = string("op_5011_cast_fp16")]; string var_5018_pad_type_0 = const()[name = string("op_5018_pad_type_0"), val = string("valid")]; tensor var_5018_strides_0 = const()[name = string("op_5018_strides_0"), val = tensor([1, 1])]; tensor var_5018_pad_0 = const()[name = string("op_5018_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5018_dilations_0 = const()[name = string("op_5018_dilations_0"), val = tensor([1, 1])]; int32 var_5018_groups_0 = const()[name = string("op_5018_groups_0"), val = int32(1)]; tensor layers_16_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_16_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(431370560)))]; tensor layers_16_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_16_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(433467776)))]; tensor var_5018_cast_fp16 = conv(bias = layers_16_self_attn_v_proj_bias_to_fp16, dilations = var_5018_dilations_0, groups = var_5018_groups_0, pad = var_5018_pad_0, pad_type = var_5018_pad_type_0, strides = var_5018_strides_0, weight = layers_16_self_attn_v_proj_weight_to_fp16, x = input_137_cast_fp16)[name = string("op_5018_cast_fp16")]; tensor tile_48 = const()[name = string("tile_48"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(433469888)))]; int32 var_5019_axis_0 = const()[name = string("op_5019_axis_0"), val = int32(1)]; tensor var_5019_cast_fp16_0, tensor var_5019_cast_fp16_1, tensor var_5019_cast_fp16_2, tensor var_5019_cast_fp16_3, tensor var_5019_cast_fp16_4, tensor var_5019_cast_fp16_5, tensor var_5019_cast_fp16_6, tensor var_5019_cast_fp16_7, tensor var_5019_cast_fp16_8, tensor var_5019_cast_fp16_9, tensor var_5019_cast_fp16_10, tensor var_5019_cast_fp16_11, tensor var_5019_cast_fp16_12, tensor var_5019_cast_fp16_13, tensor var_5019_cast_fp16_14, tensor var_5019_cast_fp16_15 = split(axis = var_5019_axis_0, split_sizes = tile_48, x = var_5004_cast_fp16)[name = string("op_5019_cast_fp16")]; tensor tile_49 = const()[name = string("tile_49"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(433470016)))]; int32 var_5036_axis_0 = const()[name = string("op_5036_axis_0"), val = int32(1)]; tensor var_5036_cast_fp16_0, tensor var_5036_cast_fp16_1, tensor var_5036_cast_fp16_2, tensor var_5036_cast_fp16_3, tensor var_5036_cast_fp16_4, tensor var_5036_cast_fp16_5, tensor var_5036_cast_fp16_6, tensor var_5036_cast_fp16_7, tensor var_5036_cast_fp16_8, tensor var_5036_cast_fp16_9, tensor var_5036_cast_fp16_10, tensor var_5036_cast_fp16_11, tensor var_5036_cast_fp16_12, tensor var_5036_cast_fp16_13, tensor var_5036_cast_fp16_14, tensor var_5036_cast_fp16_15 = split(axis = var_5036_axis_0, split_sizes = tile_49, x = var_5011_cast_fp16)[name = string("op_5036_cast_fp16")]; tensor tile_50 = const()[name = string("tile_50"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(433470144)))]; int32 var_5053_axis_0 = const()[name = string("op_5053_axis_0"), val = int32(1)]; tensor var_5053_cast_fp16_0, tensor var_5053_cast_fp16_1, tensor var_5053_cast_fp16_2, tensor var_5053_cast_fp16_3, tensor var_5053_cast_fp16_4, tensor var_5053_cast_fp16_5, tensor var_5053_cast_fp16_6, tensor var_5053_cast_fp16_7, tensor var_5053_cast_fp16_8, tensor var_5053_cast_fp16_9, tensor var_5053_cast_fp16_10, tensor var_5053_cast_fp16_11, tensor var_5053_cast_fp16_12, tensor var_5053_cast_fp16_13, tensor var_5053_cast_fp16_14, tensor var_5053_cast_fp16_15 = split(axis = var_5053_axis_0, split_sizes = tile_50, x = var_5018_cast_fp16)[name = string("op_5053_cast_fp16")]; tensor transpose_512_perm_0 = const()[name = string("transpose_512_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2564 = const()[name = string("concat_2564"), val = tensor([1, 104, 64])]; tensor transpose_512_cast_fp16 = transpose(perm = transpose_512_perm_0, x = var_5019_cast_fp16_0)[name = string("transpose_3455")]; tensor reshape_768_cast_fp16 = reshape(shape = concat_2564, x = transpose_512_cast_fp16)[name = string("reshape_768_cast_fp16")]; tensor transpose_513_perm_0 = const()[name = string("transpose_513_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2565 = const()[name = string("concat_2565"), val = tensor([1, 64, 104])]; tensor transpose_513_cast_fp16 = transpose(perm = transpose_513_perm_0, x = var_5036_cast_fp16_0)[name = string("transpose_3454")]; tensor reshape_769_cast_fp16 = reshape(shape = concat_2565, x = transpose_513_cast_fp16)[name = string("reshape_769_cast_fp16")]; bool matmul_256_transpose_x_0 = const()[name = string("matmul_256_transpose_x_0"), val = bool(false)]; bool matmul_256_transpose_y_0 = const()[name = string("matmul_256_transpose_y_0"), val = bool(false)]; tensor matmul_256_cast_fp16 = matmul(transpose_x = matmul_256_transpose_x_0, transpose_y = matmul_256_transpose_y_0, x = reshape_768_cast_fp16, y = reshape_769_cast_fp16)[name = string("matmul_256_cast_fp16")]; tensor concat_2569 = const()[name = string("concat_2569"), val = tensor([1, 1, 104, 104])]; tensor reshape_770_cast_fp16 = reshape(shape = concat_2569, x = matmul_256_cast_fp16)[name = string("reshape_770_cast_fp16")]; tensor transpose_2944_perm_0 = const()[name = string("transpose_2944_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2944 = transpose(perm = transpose_2944_perm_0, x = reshape_770_cast_fp16)[name = string("transpose_3453")]; tensor w_1027_cast_fp16 = add(x = transpose_2944, y = transpose_2305)[name = string("w_1027_cast_fp16")]; tensor var_5075_cast_fp16 = softmax(axis = var_4963, x = w_1027_cast_fp16)[name = string("op_5075_cast_fp16")]; string var_5077_equation_0 = const()[name = string("op_5077_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5077_cast_fp16 = einsum(equation = var_5077_equation_0, values = (var_5053_cast_fp16_0, var_5075_cast_fp16))[name = string("op_5077_cast_fp16")]; tensor transpose_514_perm_0 = const()[name = string("transpose_514_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2574 = const()[name = string("concat_2574"), val = tensor([1, 104, 64])]; tensor transpose_514_cast_fp16 = transpose(perm = transpose_514_perm_0, x = var_5019_cast_fp16_1)[name = string("transpose_3452")]; tensor reshape_771_cast_fp16 = reshape(shape = concat_2574, x = transpose_514_cast_fp16)[name = string("reshape_771_cast_fp16")]; tensor transpose_515_perm_0 = const()[name = string("transpose_515_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2575 = const()[name = string("concat_2575"), val = tensor([1, 64, 104])]; tensor transpose_515_cast_fp16 = transpose(perm = transpose_515_perm_0, x = var_5036_cast_fp16_1)[name = string("transpose_3451")]; tensor reshape_772_cast_fp16 = reshape(shape = concat_2575, x = transpose_515_cast_fp16)[name = string("reshape_772_cast_fp16")]; bool matmul_257_transpose_x_0 = const()[name = string("matmul_257_transpose_x_0"), val = bool(false)]; bool matmul_257_transpose_y_0 = const()[name = string("matmul_257_transpose_y_0"), val = bool(false)]; tensor matmul_257_cast_fp16 = matmul(transpose_x = matmul_257_transpose_x_0, transpose_y = matmul_257_transpose_y_0, x = reshape_771_cast_fp16, y = reshape_772_cast_fp16)[name = string("matmul_257_cast_fp16")]; tensor concat_2579 = const()[name = string("concat_2579"), val = tensor([1, 1, 104, 104])]; tensor reshape_773_cast_fp16 = reshape(shape = concat_2579, x = matmul_257_cast_fp16)[name = string("reshape_773_cast_fp16")]; tensor transpose_2945_perm_0 = const()[name = string("transpose_2945_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2945 = transpose(perm = transpose_2945_perm_0, x = reshape_773_cast_fp16)[name = string("transpose_3450")]; tensor w_1031_cast_fp16 = add(x = transpose_2945, y = transpose_2305)[name = string("w_1031_cast_fp16")]; tensor var_5083_cast_fp16 = softmax(axis = var_4963, x = w_1031_cast_fp16)[name = string("op_5083_cast_fp16")]; string var_5085_equation_0 = const()[name = string("op_5085_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5085_cast_fp16 = einsum(equation = var_5085_equation_0, values = (var_5053_cast_fp16_1, var_5083_cast_fp16))[name = string("op_5085_cast_fp16")]; tensor transpose_516_perm_0 = const()[name = string("transpose_516_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2584 = const()[name = string("concat_2584"), val = tensor([1, 104, 64])]; tensor transpose_516_cast_fp16 = transpose(perm = transpose_516_perm_0, x = var_5019_cast_fp16_2)[name = string("transpose_3449")]; tensor reshape_774_cast_fp16 = reshape(shape = concat_2584, x = transpose_516_cast_fp16)[name = string("reshape_774_cast_fp16")]; tensor transpose_517_perm_0 = const()[name = string("transpose_517_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2585 = const()[name = string("concat_2585"), val = tensor([1, 64, 104])]; tensor transpose_517_cast_fp16 = transpose(perm = transpose_517_perm_0, x = var_5036_cast_fp16_2)[name = string("transpose_3448")]; tensor reshape_775_cast_fp16 = reshape(shape = concat_2585, x = transpose_517_cast_fp16)[name = string("reshape_775_cast_fp16")]; bool matmul_258_transpose_x_0 = const()[name = string("matmul_258_transpose_x_0"), val = bool(false)]; bool matmul_258_transpose_y_0 = const()[name = string("matmul_258_transpose_y_0"), val = bool(false)]; tensor matmul_258_cast_fp16 = matmul(transpose_x = matmul_258_transpose_x_0, transpose_y = matmul_258_transpose_y_0, x = reshape_774_cast_fp16, y = reshape_775_cast_fp16)[name = string("matmul_258_cast_fp16")]; tensor concat_2589 = const()[name = string("concat_2589"), val = tensor([1, 1, 104, 104])]; tensor reshape_776_cast_fp16 = reshape(shape = concat_2589, x = matmul_258_cast_fp16)[name = string("reshape_776_cast_fp16")]; tensor transpose_2946_perm_0 = const()[name = string("transpose_2946_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2946 = transpose(perm = transpose_2946_perm_0, x = reshape_776_cast_fp16)[name = string("transpose_3447")]; tensor w_1035_cast_fp16 = add(x = transpose_2946, y = transpose_2305)[name = string("w_1035_cast_fp16")]; tensor var_5091_cast_fp16 = softmax(axis = var_4963, x = w_1035_cast_fp16)[name = string("op_5091_cast_fp16")]; string var_5093_equation_0 = const()[name = string("op_5093_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5093_cast_fp16 = einsum(equation = var_5093_equation_0, values = (var_5053_cast_fp16_2, var_5091_cast_fp16))[name = string("op_5093_cast_fp16")]; tensor transpose_518_perm_0 = const()[name = string("transpose_518_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2594 = const()[name = string("concat_2594"), val = tensor([1, 104, 64])]; tensor transpose_518_cast_fp16 = transpose(perm = transpose_518_perm_0, x = var_5019_cast_fp16_3)[name = string("transpose_3446")]; tensor reshape_777_cast_fp16 = reshape(shape = concat_2594, x = transpose_518_cast_fp16)[name = string("reshape_777_cast_fp16")]; tensor transpose_519_perm_0 = const()[name = string("transpose_519_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2595 = const()[name = string("concat_2595"), val = tensor([1, 64, 104])]; tensor transpose_519_cast_fp16 = transpose(perm = transpose_519_perm_0, x = var_5036_cast_fp16_3)[name = string("transpose_3445")]; tensor reshape_778_cast_fp16 = reshape(shape = concat_2595, x = transpose_519_cast_fp16)[name = string("reshape_778_cast_fp16")]; bool matmul_259_transpose_x_0 = const()[name = string("matmul_259_transpose_x_0"), val = bool(false)]; bool matmul_259_transpose_y_0 = const()[name = string("matmul_259_transpose_y_0"), val = bool(false)]; tensor matmul_259_cast_fp16 = matmul(transpose_x = matmul_259_transpose_x_0, transpose_y = matmul_259_transpose_y_0, x = reshape_777_cast_fp16, y = reshape_778_cast_fp16)[name = string("matmul_259_cast_fp16")]; tensor concat_2599 = const()[name = string("concat_2599"), val = tensor([1, 1, 104, 104])]; tensor reshape_779_cast_fp16 = reshape(shape = concat_2599, x = matmul_259_cast_fp16)[name = string("reshape_779_cast_fp16")]; tensor transpose_2947_perm_0 = const()[name = string("transpose_2947_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2947 = transpose(perm = transpose_2947_perm_0, x = reshape_779_cast_fp16)[name = string("transpose_3444")]; tensor w_1039_cast_fp16 = add(x = transpose_2947, y = transpose_2305)[name = string("w_1039_cast_fp16")]; tensor var_5099_cast_fp16 = softmax(axis = var_4963, x = w_1039_cast_fp16)[name = string("op_5099_cast_fp16")]; string var_5101_equation_0 = const()[name = string("op_5101_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5101_cast_fp16 = einsum(equation = var_5101_equation_0, values = (var_5053_cast_fp16_3, var_5099_cast_fp16))[name = string("op_5101_cast_fp16")]; tensor transpose_520_perm_0 = const()[name = string("transpose_520_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2604 = const()[name = string("concat_2604"), val = tensor([1, 104, 64])]; tensor transpose_520_cast_fp16 = transpose(perm = transpose_520_perm_0, x = var_5019_cast_fp16_4)[name = string("transpose_3443")]; tensor reshape_780_cast_fp16 = reshape(shape = concat_2604, x = transpose_520_cast_fp16)[name = string("reshape_780_cast_fp16")]; tensor transpose_521_perm_0 = const()[name = string("transpose_521_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2605 = const()[name = string("concat_2605"), val = tensor([1, 64, 104])]; tensor transpose_521_cast_fp16 = transpose(perm = transpose_521_perm_0, x = var_5036_cast_fp16_4)[name = string("transpose_3442")]; tensor reshape_781_cast_fp16 = reshape(shape = concat_2605, x = transpose_521_cast_fp16)[name = string("reshape_781_cast_fp16")]; bool matmul_260_transpose_x_0 = const()[name = string("matmul_260_transpose_x_0"), val = bool(false)]; bool matmul_260_transpose_y_0 = const()[name = string("matmul_260_transpose_y_0"), val = bool(false)]; tensor matmul_260_cast_fp16 = matmul(transpose_x = matmul_260_transpose_x_0, transpose_y = matmul_260_transpose_y_0, x = reshape_780_cast_fp16, y = reshape_781_cast_fp16)[name = string("matmul_260_cast_fp16")]; tensor concat_2609 = const()[name = string("concat_2609"), val = tensor([1, 1, 104, 104])]; tensor reshape_782_cast_fp16 = reshape(shape = concat_2609, x = matmul_260_cast_fp16)[name = string("reshape_782_cast_fp16")]; tensor transpose_2948_perm_0 = const()[name = string("transpose_2948_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2948 = transpose(perm = transpose_2948_perm_0, x = reshape_782_cast_fp16)[name = string("transpose_3441")]; tensor w_1043_cast_fp16 = add(x = transpose_2948, y = transpose_2305)[name = string("w_1043_cast_fp16")]; tensor var_5107_cast_fp16 = softmax(axis = var_4963, x = w_1043_cast_fp16)[name = string("op_5107_cast_fp16")]; string var_5109_equation_0 = const()[name = string("op_5109_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5109_cast_fp16 = einsum(equation = var_5109_equation_0, values = (var_5053_cast_fp16_4, var_5107_cast_fp16))[name = string("op_5109_cast_fp16")]; tensor transpose_522_perm_0 = const()[name = string("transpose_522_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2614 = const()[name = string("concat_2614"), val = tensor([1, 104, 64])]; tensor transpose_522_cast_fp16 = transpose(perm = transpose_522_perm_0, x = var_5019_cast_fp16_5)[name = string("transpose_3440")]; tensor reshape_783_cast_fp16 = reshape(shape = concat_2614, x = transpose_522_cast_fp16)[name = string("reshape_783_cast_fp16")]; tensor transpose_523_perm_0 = const()[name = string("transpose_523_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2615 = const()[name = string("concat_2615"), val = tensor([1, 64, 104])]; tensor transpose_523_cast_fp16 = transpose(perm = transpose_523_perm_0, x = var_5036_cast_fp16_5)[name = string("transpose_3439")]; tensor reshape_784_cast_fp16 = reshape(shape = concat_2615, x = transpose_523_cast_fp16)[name = string("reshape_784_cast_fp16")]; bool matmul_261_transpose_x_0 = const()[name = string("matmul_261_transpose_x_0"), val = bool(false)]; bool matmul_261_transpose_y_0 = const()[name = string("matmul_261_transpose_y_0"), val = bool(false)]; tensor matmul_261_cast_fp16 = matmul(transpose_x = matmul_261_transpose_x_0, transpose_y = matmul_261_transpose_y_0, x = reshape_783_cast_fp16, y = reshape_784_cast_fp16)[name = string("matmul_261_cast_fp16")]; tensor concat_2619 = const()[name = string("concat_2619"), val = tensor([1, 1, 104, 104])]; tensor reshape_785_cast_fp16 = reshape(shape = concat_2619, x = matmul_261_cast_fp16)[name = string("reshape_785_cast_fp16")]; tensor transpose_2949_perm_0 = const()[name = string("transpose_2949_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2949 = transpose(perm = transpose_2949_perm_0, x = reshape_785_cast_fp16)[name = string("transpose_3438")]; tensor w_1047_cast_fp16 = add(x = transpose_2949, y = transpose_2305)[name = string("w_1047_cast_fp16")]; tensor var_5115_cast_fp16 = softmax(axis = var_4963, x = w_1047_cast_fp16)[name = string("op_5115_cast_fp16")]; string var_5117_equation_0 = const()[name = string("op_5117_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5117_cast_fp16 = einsum(equation = var_5117_equation_0, values = (var_5053_cast_fp16_5, var_5115_cast_fp16))[name = string("op_5117_cast_fp16")]; tensor transpose_524_perm_0 = const()[name = string("transpose_524_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2624 = const()[name = string("concat_2624"), val = tensor([1, 104, 64])]; tensor transpose_524_cast_fp16 = transpose(perm = transpose_524_perm_0, x = var_5019_cast_fp16_6)[name = string("transpose_3437")]; tensor reshape_786_cast_fp16 = reshape(shape = concat_2624, x = transpose_524_cast_fp16)[name = string("reshape_786_cast_fp16")]; tensor transpose_525_perm_0 = const()[name = string("transpose_525_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2625 = const()[name = string("concat_2625"), val = tensor([1, 64, 104])]; tensor transpose_525_cast_fp16 = transpose(perm = transpose_525_perm_0, x = var_5036_cast_fp16_6)[name = string("transpose_3436")]; tensor reshape_787_cast_fp16 = reshape(shape = concat_2625, x = transpose_525_cast_fp16)[name = string("reshape_787_cast_fp16")]; bool matmul_262_transpose_x_0 = const()[name = string("matmul_262_transpose_x_0"), val = bool(false)]; bool matmul_262_transpose_y_0 = const()[name = string("matmul_262_transpose_y_0"), val = bool(false)]; tensor matmul_262_cast_fp16 = matmul(transpose_x = matmul_262_transpose_x_0, transpose_y = matmul_262_transpose_y_0, x = reshape_786_cast_fp16, y = reshape_787_cast_fp16)[name = string("matmul_262_cast_fp16")]; tensor concat_2629 = const()[name = string("concat_2629"), val = tensor([1, 1, 104, 104])]; tensor reshape_788_cast_fp16 = reshape(shape = concat_2629, x = matmul_262_cast_fp16)[name = string("reshape_788_cast_fp16")]; tensor transpose_2950_perm_0 = const()[name = string("transpose_2950_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2950 = transpose(perm = transpose_2950_perm_0, x = reshape_788_cast_fp16)[name = string("transpose_3435")]; tensor w_1051_cast_fp16 = add(x = transpose_2950, y = transpose_2305)[name = string("w_1051_cast_fp16")]; tensor var_5123_cast_fp16 = softmax(axis = var_4963, x = w_1051_cast_fp16)[name = string("op_5123_cast_fp16")]; string var_5125_equation_0 = const()[name = string("op_5125_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5125_cast_fp16 = einsum(equation = var_5125_equation_0, values = (var_5053_cast_fp16_6, var_5123_cast_fp16))[name = string("op_5125_cast_fp16")]; tensor transpose_526_perm_0 = const()[name = string("transpose_526_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2634 = const()[name = string("concat_2634"), val = tensor([1, 104, 64])]; tensor transpose_526_cast_fp16 = transpose(perm = transpose_526_perm_0, x = var_5019_cast_fp16_7)[name = string("transpose_3434")]; tensor reshape_789_cast_fp16 = reshape(shape = concat_2634, x = transpose_526_cast_fp16)[name = string("reshape_789_cast_fp16")]; tensor transpose_527_perm_0 = const()[name = string("transpose_527_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2635 = const()[name = string("concat_2635"), val = tensor([1, 64, 104])]; tensor transpose_527_cast_fp16 = transpose(perm = transpose_527_perm_0, x = var_5036_cast_fp16_7)[name = string("transpose_3433")]; tensor reshape_790_cast_fp16 = reshape(shape = concat_2635, x = transpose_527_cast_fp16)[name = string("reshape_790_cast_fp16")]; bool matmul_263_transpose_x_0 = const()[name = string("matmul_263_transpose_x_0"), val = bool(false)]; bool matmul_263_transpose_y_0 = const()[name = string("matmul_263_transpose_y_0"), val = bool(false)]; tensor matmul_263_cast_fp16 = matmul(transpose_x = matmul_263_transpose_x_0, transpose_y = matmul_263_transpose_y_0, x = reshape_789_cast_fp16, y = reshape_790_cast_fp16)[name = string("matmul_263_cast_fp16")]; tensor concat_2639 = const()[name = string("concat_2639"), val = tensor([1, 1, 104, 104])]; tensor reshape_791_cast_fp16 = reshape(shape = concat_2639, x = matmul_263_cast_fp16)[name = string("reshape_791_cast_fp16")]; tensor transpose_2951_perm_0 = const()[name = string("transpose_2951_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2951 = transpose(perm = transpose_2951_perm_0, x = reshape_791_cast_fp16)[name = string("transpose_3432")]; tensor w_1055_cast_fp16 = add(x = transpose_2951, y = transpose_2305)[name = string("w_1055_cast_fp16")]; tensor var_5131_cast_fp16 = softmax(axis = var_4963, x = w_1055_cast_fp16)[name = string("op_5131_cast_fp16")]; string var_5133_equation_0 = const()[name = string("op_5133_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5133_cast_fp16 = einsum(equation = var_5133_equation_0, values = (var_5053_cast_fp16_7, var_5131_cast_fp16))[name = string("op_5133_cast_fp16")]; tensor transpose_528_perm_0 = const()[name = string("transpose_528_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2644 = const()[name = string("concat_2644"), val = tensor([1, 104, 64])]; tensor transpose_528_cast_fp16 = transpose(perm = transpose_528_perm_0, x = var_5019_cast_fp16_8)[name = string("transpose_3431")]; tensor reshape_792_cast_fp16 = reshape(shape = concat_2644, x = transpose_528_cast_fp16)[name = string("reshape_792_cast_fp16")]; tensor transpose_529_perm_0 = const()[name = string("transpose_529_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2645 = const()[name = string("concat_2645"), val = tensor([1, 64, 104])]; tensor transpose_529_cast_fp16 = transpose(perm = transpose_529_perm_0, x = var_5036_cast_fp16_8)[name = string("transpose_3430")]; tensor reshape_793_cast_fp16 = reshape(shape = concat_2645, x = transpose_529_cast_fp16)[name = string("reshape_793_cast_fp16")]; bool matmul_264_transpose_x_0 = const()[name = string("matmul_264_transpose_x_0"), val = bool(false)]; bool matmul_264_transpose_y_0 = const()[name = string("matmul_264_transpose_y_0"), val = bool(false)]; tensor matmul_264_cast_fp16 = matmul(transpose_x = matmul_264_transpose_x_0, transpose_y = matmul_264_transpose_y_0, x = reshape_792_cast_fp16, y = reshape_793_cast_fp16)[name = string("matmul_264_cast_fp16")]; tensor concat_2649 = const()[name = string("concat_2649"), val = tensor([1, 1, 104, 104])]; tensor reshape_794_cast_fp16 = reshape(shape = concat_2649, x = matmul_264_cast_fp16)[name = string("reshape_794_cast_fp16")]; tensor transpose_2952_perm_0 = const()[name = string("transpose_2952_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2952 = transpose(perm = transpose_2952_perm_0, x = reshape_794_cast_fp16)[name = string("transpose_3429")]; tensor w_1059_cast_fp16 = add(x = transpose_2952, y = transpose_2305)[name = string("w_1059_cast_fp16")]; tensor var_5139_cast_fp16 = softmax(axis = var_4963, x = w_1059_cast_fp16)[name = string("op_5139_cast_fp16")]; string var_5141_equation_0 = const()[name = string("op_5141_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5141_cast_fp16 = einsum(equation = var_5141_equation_0, values = (var_5053_cast_fp16_8, var_5139_cast_fp16))[name = string("op_5141_cast_fp16")]; tensor transpose_530_perm_0 = const()[name = string("transpose_530_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2654 = const()[name = string("concat_2654"), val = tensor([1, 104, 64])]; tensor transpose_530_cast_fp16 = transpose(perm = transpose_530_perm_0, x = var_5019_cast_fp16_9)[name = string("transpose_3428")]; tensor reshape_795_cast_fp16 = reshape(shape = concat_2654, x = transpose_530_cast_fp16)[name = string("reshape_795_cast_fp16")]; tensor transpose_531_perm_0 = const()[name = string("transpose_531_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2655 = const()[name = string("concat_2655"), val = tensor([1, 64, 104])]; tensor transpose_531_cast_fp16 = transpose(perm = transpose_531_perm_0, x = var_5036_cast_fp16_9)[name = string("transpose_3427")]; tensor reshape_796_cast_fp16 = reshape(shape = concat_2655, x = transpose_531_cast_fp16)[name = string("reshape_796_cast_fp16")]; bool matmul_265_transpose_x_0 = const()[name = string("matmul_265_transpose_x_0"), val = bool(false)]; bool matmul_265_transpose_y_0 = const()[name = string("matmul_265_transpose_y_0"), val = bool(false)]; tensor matmul_265_cast_fp16 = matmul(transpose_x = matmul_265_transpose_x_0, transpose_y = matmul_265_transpose_y_0, x = reshape_795_cast_fp16, y = reshape_796_cast_fp16)[name = string("matmul_265_cast_fp16")]; tensor concat_2659 = const()[name = string("concat_2659"), val = tensor([1, 1, 104, 104])]; tensor reshape_797_cast_fp16 = reshape(shape = concat_2659, x = matmul_265_cast_fp16)[name = string("reshape_797_cast_fp16")]; tensor transpose_2953_perm_0 = const()[name = string("transpose_2953_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2953 = transpose(perm = transpose_2953_perm_0, x = reshape_797_cast_fp16)[name = string("transpose_3426")]; tensor w_1063_cast_fp16 = add(x = transpose_2953, y = transpose_2305)[name = string("w_1063_cast_fp16")]; tensor var_5147_cast_fp16 = softmax(axis = var_4963, x = w_1063_cast_fp16)[name = string("op_5147_cast_fp16")]; string var_5149_equation_0 = const()[name = string("op_5149_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5149_cast_fp16 = einsum(equation = var_5149_equation_0, values = (var_5053_cast_fp16_9, var_5147_cast_fp16))[name = string("op_5149_cast_fp16")]; tensor transpose_532_perm_0 = const()[name = string("transpose_532_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2664 = const()[name = string("concat_2664"), val = tensor([1, 104, 64])]; tensor transpose_532_cast_fp16 = transpose(perm = transpose_532_perm_0, x = var_5019_cast_fp16_10)[name = string("transpose_3425")]; tensor reshape_798_cast_fp16 = reshape(shape = concat_2664, x = transpose_532_cast_fp16)[name = string("reshape_798_cast_fp16")]; tensor transpose_533_perm_0 = const()[name = string("transpose_533_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2665 = const()[name = string("concat_2665"), val = tensor([1, 64, 104])]; tensor transpose_533_cast_fp16 = transpose(perm = transpose_533_perm_0, x = var_5036_cast_fp16_10)[name = string("transpose_3424")]; tensor reshape_799_cast_fp16 = reshape(shape = concat_2665, x = transpose_533_cast_fp16)[name = string("reshape_799_cast_fp16")]; bool matmul_266_transpose_x_0 = const()[name = string("matmul_266_transpose_x_0"), val = bool(false)]; bool matmul_266_transpose_y_0 = const()[name = string("matmul_266_transpose_y_0"), val = bool(false)]; tensor matmul_266_cast_fp16 = matmul(transpose_x = matmul_266_transpose_x_0, transpose_y = matmul_266_transpose_y_0, x = reshape_798_cast_fp16, y = reshape_799_cast_fp16)[name = string("matmul_266_cast_fp16")]; tensor concat_2669 = const()[name = string("concat_2669"), val = tensor([1, 1, 104, 104])]; tensor reshape_800_cast_fp16 = reshape(shape = concat_2669, x = matmul_266_cast_fp16)[name = string("reshape_800_cast_fp16")]; tensor transpose_2954_perm_0 = const()[name = string("transpose_2954_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2954 = transpose(perm = transpose_2954_perm_0, x = reshape_800_cast_fp16)[name = string("transpose_3423")]; tensor w_1067_cast_fp16 = add(x = transpose_2954, y = transpose_2305)[name = string("w_1067_cast_fp16")]; tensor var_5155_cast_fp16 = softmax(axis = var_4963, x = w_1067_cast_fp16)[name = string("op_5155_cast_fp16")]; string var_5157_equation_0 = const()[name = string("op_5157_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5157_cast_fp16 = einsum(equation = var_5157_equation_0, values = (var_5053_cast_fp16_10, var_5155_cast_fp16))[name = string("op_5157_cast_fp16")]; tensor transpose_534_perm_0 = const()[name = string("transpose_534_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2674 = const()[name = string("concat_2674"), val = tensor([1, 104, 64])]; tensor transpose_534_cast_fp16 = transpose(perm = transpose_534_perm_0, x = var_5019_cast_fp16_11)[name = string("transpose_3422")]; tensor reshape_801_cast_fp16 = reshape(shape = concat_2674, x = transpose_534_cast_fp16)[name = string("reshape_801_cast_fp16")]; tensor transpose_535_perm_0 = const()[name = string("transpose_535_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2675 = const()[name = string("concat_2675"), val = tensor([1, 64, 104])]; tensor transpose_535_cast_fp16 = transpose(perm = transpose_535_perm_0, x = var_5036_cast_fp16_11)[name = string("transpose_3421")]; tensor reshape_802_cast_fp16 = reshape(shape = concat_2675, x = transpose_535_cast_fp16)[name = string("reshape_802_cast_fp16")]; bool matmul_267_transpose_x_0 = const()[name = string("matmul_267_transpose_x_0"), val = bool(false)]; bool matmul_267_transpose_y_0 = const()[name = string("matmul_267_transpose_y_0"), val = bool(false)]; tensor matmul_267_cast_fp16 = matmul(transpose_x = matmul_267_transpose_x_0, transpose_y = matmul_267_transpose_y_0, x = reshape_801_cast_fp16, y = reshape_802_cast_fp16)[name = string("matmul_267_cast_fp16")]; tensor concat_2679 = const()[name = string("concat_2679"), val = tensor([1, 1, 104, 104])]; tensor reshape_803_cast_fp16 = reshape(shape = concat_2679, x = matmul_267_cast_fp16)[name = string("reshape_803_cast_fp16")]; tensor transpose_2955_perm_0 = const()[name = string("transpose_2955_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2955 = transpose(perm = transpose_2955_perm_0, x = reshape_803_cast_fp16)[name = string("transpose_3420")]; tensor w_1071_cast_fp16 = add(x = transpose_2955, y = transpose_2305)[name = string("w_1071_cast_fp16")]; tensor var_5163_cast_fp16 = softmax(axis = var_4963, x = w_1071_cast_fp16)[name = string("op_5163_cast_fp16")]; string var_5165_equation_0 = const()[name = string("op_5165_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5165_cast_fp16 = einsum(equation = var_5165_equation_0, values = (var_5053_cast_fp16_11, var_5163_cast_fp16))[name = string("op_5165_cast_fp16")]; tensor transpose_536_perm_0 = const()[name = string("transpose_536_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2684 = const()[name = string("concat_2684"), val = tensor([1, 104, 64])]; tensor transpose_536_cast_fp16 = transpose(perm = transpose_536_perm_0, x = var_5019_cast_fp16_12)[name = string("transpose_3419")]; tensor reshape_804_cast_fp16 = reshape(shape = concat_2684, x = transpose_536_cast_fp16)[name = string("reshape_804_cast_fp16")]; tensor transpose_537_perm_0 = const()[name = string("transpose_537_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2685 = const()[name = string("concat_2685"), val = tensor([1, 64, 104])]; tensor transpose_537_cast_fp16 = transpose(perm = transpose_537_perm_0, x = var_5036_cast_fp16_12)[name = string("transpose_3418")]; tensor reshape_805_cast_fp16 = reshape(shape = concat_2685, x = transpose_537_cast_fp16)[name = string("reshape_805_cast_fp16")]; bool matmul_268_transpose_x_0 = const()[name = string("matmul_268_transpose_x_0"), val = bool(false)]; bool matmul_268_transpose_y_0 = const()[name = string("matmul_268_transpose_y_0"), val = bool(false)]; tensor matmul_268_cast_fp16 = matmul(transpose_x = matmul_268_transpose_x_0, transpose_y = matmul_268_transpose_y_0, x = reshape_804_cast_fp16, y = reshape_805_cast_fp16)[name = string("matmul_268_cast_fp16")]; tensor concat_2689 = const()[name = string("concat_2689"), val = tensor([1, 1, 104, 104])]; tensor reshape_806_cast_fp16 = reshape(shape = concat_2689, x = matmul_268_cast_fp16)[name = string("reshape_806_cast_fp16")]; tensor transpose_2956_perm_0 = const()[name = string("transpose_2956_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2956 = transpose(perm = transpose_2956_perm_0, x = reshape_806_cast_fp16)[name = string("transpose_3417")]; tensor w_1075_cast_fp16 = add(x = transpose_2956, y = transpose_2305)[name = string("w_1075_cast_fp16")]; tensor var_5171_cast_fp16 = softmax(axis = var_4963, x = w_1075_cast_fp16)[name = string("op_5171_cast_fp16")]; string var_5173_equation_0 = const()[name = string("op_5173_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5173_cast_fp16 = einsum(equation = var_5173_equation_0, values = (var_5053_cast_fp16_12, var_5171_cast_fp16))[name = string("op_5173_cast_fp16")]; tensor transpose_538_perm_0 = const()[name = string("transpose_538_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2694 = const()[name = string("concat_2694"), val = tensor([1, 104, 64])]; tensor transpose_538_cast_fp16 = transpose(perm = transpose_538_perm_0, x = var_5019_cast_fp16_13)[name = string("transpose_3416")]; tensor reshape_807_cast_fp16 = reshape(shape = concat_2694, x = transpose_538_cast_fp16)[name = string("reshape_807_cast_fp16")]; tensor transpose_539_perm_0 = const()[name = string("transpose_539_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2695 = const()[name = string("concat_2695"), val = tensor([1, 64, 104])]; tensor transpose_539_cast_fp16 = transpose(perm = transpose_539_perm_0, x = var_5036_cast_fp16_13)[name = string("transpose_3415")]; tensor reshape_808_cast_fp16 = reshape(shape = concat_2695, x = transpose_539_cast_fp16)[name = string("reshape_808_cast_fp16")]; bool matmul_269_transpose_x_0 = const()[name = string("matmul_269_transpose_x_0"), val = bool(false)]; bool matmul_269_transpose_y_0 = const()[name = string("matmul_269_transpose_y_0"), val = bool(false)]; tensor matmul_269_cast_fp16 = matmul(transpose_x = matmul_269_transpose_x_0, transpose_y = matmul_269_transpose_y_0, x = reshape_807_cast_fp16, y = reshape_808_cast_fp16)[name = string("matmul_269_cast_fp16")]; tensor concat_2699 = const()[name = string("concat_2699"), val = tensor([1, 1, 104, 104])]; tensor reshape_809_cast_fp16 = reshape(shape = concat_2699, x = matmul_269_cast_fp16)[name = string("reshape_809_cast_fp16")]; tensor transpose_2957_perm_0 = const()[name = string("transpose_2957_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2957 = transpose(perm = transpose_2957_perm_0, x = reshape_809_cast_fp16)[name = string("transpose_3414")]; tensor w_1079_cast_fp16 = add(x = transpose_2957, y = transpose_2305)[name = string("w_1079_cast_fp16")]; tensor var_5179_cast_fp16 = softmax(axis = var_4963, x = w_1079_cast_fp16)[name = string("op_5179_cast_fp16")]; string var_5181_equation_0 = const()[name = string("op_5181_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5181_cast_fp16 = einsum(equation = var_5181_equation_0, values = (var_5053_cast_fp16_13, var_5179_cast_fp16))[name = string("op_5181_cast_fp16")]; tensor transpose_540_perm_0 = const()[name = string("transpose_540_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2704 = const()[name = string("concat_2704"), val = tensor([1, 104, 64])]; tensor transpose_540_cast_fp16 = transpose(perm = transpose_540_perm_0, x = var_5019_cast_fp16_14)[name = string("transpose_3413")]; tensor reshape_810_cast_fp16 = reshape(shape = concat_2704, x = transpose_540_cast_fp16)[name = string("reshape_810_cast_fp16")]; tensor transpose_541_perm_0 = const()[name = string("transpose_541_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2705 = const()[name = string("concat_2705"), val = tensor([1, 64, 104])]; tensor transpose_541_cast_fp16 = transpose(perm = transpose_541_perm_0, x = var_5036_cast_fp16_14)[name = string("transpose_3412")]; tensor reshape_811_cast_fp16 = reshape(shape = concat_2705, x = transpose_541_cast_fp16)[name = string("reshape_811_cast_fp16")]; bool matmul_270_transpose_x_0 = const()[name = string("matmul_270_transpose_x_0"), val = bool(false)]; bool matmul_270_transpose_y_0 = const()[name = string("matmul_270_transpose_y_0"), val = bool(false)]; tensor matmul_270_cast_fp16 = matmul(transpose_x = matmul_270_transpose_x_0, transpose_y = matmul_270_transpose_y_0, x = reshape_810_cast_fp16, y = reshape_811_cast_fp16)[name = string("matmul_270_cast_fp16")]; tensor concat_2709 = const()[name = string("concat_2709"), val = tensor([1, 1, 104, 104])]; tensor reshape_812_cast_fp16 = reshape(shape = concat_2709, x = matmul_270_cast_fp16)[name = string("reshape_812_cast_fp16")]; tensor transpose_2958_perm_0 = const()[name = string("transpose_2958_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2958 = transpose(perm = transpose_2958_perm_0, x = reshape_812_cast_fp16)[name = string("transpose_3411")]; tensor w_1083_cast_fp16 = add(x = transpose_2958, y = transpose_2305)[name = string("w_1083_cast_fp16")]; tensor var_5187_cast_fp16 = softmax(axis = var_4963, x = w_1083_cast_fp16)[name = string("op_5187_cast_fp16")]; string var_5189_equation_0 = const()[name = string("op_5189_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5189_cast_fp16 = einsum(equation = var_5189_equation_0, values = (var_5053_cast_fp16_14, var_5187_cast_fp16))[name = string("op_5189_cast_fp16")]; tensor transpose_542_perm_0 = const()[name = string("transpose_542_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2714 = const()[name = string("concat_2714"), val = tensor([1, 104, 64])]; tensor transpose_542_cast_fp16 = transpose(perm = transpose_542_perm_0, x = var_5019_cast_fp16_15)[name = string("transpose_3410")]; tensor reshape_813_cast_fp16 = reshape(shape = concat_2714, x = transpose_542_cast_fp16)[name = string("reshape_813_cast_fp16")]; tensor transpose_543_perm_0 = const()[name = string("transpose_543_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2715 = const()[name = string("concat_2715"), val = tensor([1, 64, 104])]; tensor transpose_543_cast_fp16 = transpose(perm = transpose_543_perm_0, x = var_5036_cast_fp16_15)[name = string("transpose_3409")]; tensor reshape_814_cast_fp16 = reshape(shape = concat_2715, x = transpose_543_cast_fp16)[name = string("reshape_814_cast_fp16")]; bool matmul_271_transpose_x_0 = const()[name = string("matmul_271_transpose_x_0"), val = bool(false)]; bool matmul_271_transpose_y_0 = const()[name = string("matmul_271_transpose_y_0"), val = bool(false)]; tensor matmul_271_cast_fp16 = matmul(transpose_x = matmul_271_transpose_x_0, transpose_y = matmul_271_transpose_y_0, x = reshape_813_cast_fp16, y = reshape_814_cast_fp16)[name = string("matmul_271_cast_fp16")]; tensor concat_2719 = const()[name = string("concat_2719"), val = tensor([1, 1, 104, 104])]; tensor reshape_815_cast_fp16 = reshape(shape = concat_2719, x = matmul_271_cast_fp16)[name = string("reshape_815_cast_fp16")]; tensor transpose_2959_perm_0 = const()[name = string("transpose_2959_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2959 = transpose(perm = transpose_2959_perm_0, x = reshape_815_cast_fp16)[name = string("transpose_3408")]; tensor w_1087_cast_fp16 = add(x = transpose_2959, y = transpose_2305)[name = string("w_1087_cast_fp16")]; tensor var_5195_cast_fp16 = softmax(axis = var_4963, x = w_1087_cast_fp16)[name = string("op_5195_cast_fp16")]; string var_5197_equation_0 = const()[name = string("op_5197_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5197_cast_fp16 = einsum(equation = var_5197_equation_0, values = (var_5053_cast_fp16_15, var_5195_cast_fp16))[name = string("op_5197_cast_fp16")]; bool input_139_interleave_0 = const()[name = string("input_139_interleave_0"), val = bool(false)]; tensor input_139_cast_fp16 = concat(axis = var_4963, interleave = input_139_interleave_0, values = (var_5077_cast_fp16, var_5085_cast_fp16, var_5093_cast_fp16, var_5101_cast_fp16, var_5109_cast_fp16, var_5117_cast_fp16, var_5125_cast_fp16, var_5133_cast_fp16, var_5141_cast_fp16, var_5149_cast_fp16, var_5157_cast_fp16, var_5165_cast_fp16, var_5173_cast_fp16, var_5181_cast_fp16, var_5189_cast_fp16, var_5197_cast_fp16))[name = string("input_139_cast_fp16")]; string var_5206_pad_type_0 = const()[name = string("op_5206_pad_type_0"), val = string("valid")]; tensor var_5206_strides_0 = const()[name = string("op_5206_strides_0"), val = tensor([1, 1])]; tensor var_5206_pad_0 = const()[name = string("op_5206_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5206_dilations_0 = const()[name = string("op_5206_dilations_0"), val = tensor([1, 1])]; int32 var_5206_groups_0 = const()[name = string("op_5206_groups_0"), val = int32(1)]; tensor layers_16_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_16_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(433470272)))]; tensor layers_16_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_16_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(435567488)))]; tensor var_5206_cast_fp16 = conv(bias = layers_16_self_attn_out_proj_bias_to_fp16, dilations = var_5206_dilations_0, groups = var_5206_groups_0, pad = var_5206_pad_0, pad_type = var_5206_pad_type_0, strides = var_5206_strides_0, weight = layers_16_self_attn_out_proj_weight_to_fp16, x = input_139_cast_fp16)[name = string("op_5206_cast_fp16")]; tensor x_177_cast_fp16 = add(x = x_173_cast_fp16, y = var_5206_cast_fp16)[name = string("x_177_cast_fp16")]; tensor mu_67_axes_0 = const()[name = string("mu_67_axes_0"), val = tensor([1])]; bool mu_67_keep_dims_0 = const()[name = string("mu_67_keep_dims_0"), val = bool(true)]; tensor mu_67_cast_fp16 = reduce_mean(axes = mu_67_axes_0, keep_dims = mu_67_keep_dims_0, x = x_177_cast_fp16)[name = string("mu_67_cast_fp16")]; tensor var_5212_cast_fp16 = sub(x = x_177_cast_fp16, y = mu_67_cast_fp16)[name = string("op_5212_cast_fp16")]; fp16 var_4966_promoted_1_to_fp16 = const()[name = string("op_4966_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_5213_cast_fp16 = pow(x = var_5212_cast_fp16, y = var_4966_promoted_1_to_fp16)[name = string("op_5213_cast_fp16")]; tensor var_67_axes_0 = const()[name = string("var_67_axes_0"), val = tensor([1])]; bool var_67_keep_dims_0 = const()[name = string("var_67_keep_dims_0"), val = bool(true)]; tensor var_67_cast_fp16 = reduce_mean(axes = var_67_axes_0, keep_dims = var_67_keep_dims_0, x = var_5213_cast_fp16)[name = string("var_67_cast_fp16")]; fp16 var_5217_to_fp16 = const()[name = string("op_5217_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_5218_cast_fp16 = add(x = var_67_cast_fp16, y = var_5217_to_fp16)[name = string("op_5218_cast_fp16")]; fp32 var_5219_epsilon_0 = const()[name = string("op_5219_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_5219_cast_fp16 = rsqrt(epsilon = var_5219_epsilon_0, x = var_5218_cast_fp16)[name = string("op_5219_cast_fp16")]; tensor x_179_cast_fp16 = mul(x = var_5212_cast_fp16, y = var_5219_cast_fp16)[name = string("x_179_cast_fp16")]; tensor input_141_gamma_0_to_fp16 = const()[name = string("input_141_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(435569600)))]; tensor input_141_beta_0_to_fp16 = const()[name = string("input_141_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(435571712)))]; fp16 input_141_epsilon_0_to_fp16 = const()[name = string("input_141_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_141_cast_fp16 = batch_norm(beta = input_141_beta_0_to_fp16, epsilon = input_141_epsilon_0_to_fp16, gamma = input_141_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_179_cast_fp16)[name = string("input_141_cast_fp16")]; string x_181_pad_type_0 = const()[name = string("x_181_pad_type_0"), val = string("valid")]; tensor x_181_strides_0 = const()[name = string("x_181_strides_0"), val = tensor([1, 1])]; tensor x_181_pad_0 = const()[name = string("x_181_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_181_dilations_0 = const()[name = string("x_181_dilations_0"), val = tensor([1, 1])]; int32 x_181_groups_0 = const()[name = string("x_181_groups_0"), val = int32(1)]; tensor layers_16_fc1_weight_to_fp16 = const()[name = string("layers_16_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(435573824)))]; tensor layers_16_fc1_bias_to_fp16 = const()[name = string("layers_16_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(443962496)))]; tensor x_181_cast_fp16 = conv(bias = layers_16_fc1_bias_to_fp16, dilations = x_181_dilations_0, groups = x_181_groups_0, pad = x_181_pad_0, pad_type = x_181_pad_type_0, strides = x_181_strides_0, weight = layers_16_fc1_weight_to_fp16, x = input_141_cast_fp16)[name = string("x_181_cast_fp16")]; fp16 var_5234_to_fp16 = const()[name = string("op_5234_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_5235_cast_fp16 = mul(x = x_181_cast_fp16, y = var_5234_to_fp16)[name = string("op_5235_cast_fp16")]; tensor var_5236_cast_fp16 = mul(x = var_5235_cast_fp16, y = x_181_cast_fp16)[name = string("op_5236_cast_fp16")]; tensor var_5237_cast_fp16 = mul(x = var_5236_cast_fp16, y = x_181_cast_fp16)[name = string("op_5237_cast_fp16")]; tensor var_5238_cast_fp16 = add(x = x_181_cast_fp16, y = var_5237_cast_fp16)[name = string("op_5238_cast_fp16")]; fp16 var_5239_to_fp16 = const()[name = string("op_5239_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_39_cast_fp16 = mul(x = var_5238_cast_fp16, y = var_5239_to_fp16)[name = string("u_39_cast_fp16")]; fp16 var_5241_to_fp16 = const()[name = string("op_5241_to_fp16"), val = fp16(0x1p-1)]; tensor var_5242_cast_fp16 = mul(x = x_181_cast_fp16, y = var_5241_to_fp16)[name = string("op_5242_cast_fp16")]; tensor var_5243_cast_fp16 = tanh(x = u_39_cast_fp16)[name = string("op_5243_cast_fp16")]; fp16 var_5244_to_fp16 = const()[name = string("op_5244_to_fp16"), val = fp16(0x1p+0)]; tensor var_5245_cast_fp16 = add(x = var_5243_cast_fp16, y = var_5244_to_fp16)[name = string("op_5245_cast_fp16")]; tensor input_143_cast_fp16 = mul(x = var_5242_cast_fp16, y = var_5245_cast_fp16)[name = string("input_143_cast_fp16")]; string h_33_pad_type_0 = const()[name = string("h_33_pad_type_0"), val = string("valid")]; tensor h_33_strides_0 = const()[name = string("h_33_strides_0"), val = tensor([1, 1])]; tensor h_33_pad_0 = const()[name = string("h_33_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_33_dilations_0 = const()[name = string("h_33_dilations_0"), val = tensor([1, 1])]; int32 h_33_groups_0 = const()[name = string("h_33_groups_0"), val = int32(1)]; tensor layers_16_fc2_weight_to_fp16 = const()[name = string("layers_16_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(443970752)))]; tensor layers_16_fc2_bias_to_fp16 = const()[name = string("layers_16_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(452359424)))]; tensor h_33_cast_fp16 = conv(bias = layers_16_fc2_bias_to_fp16, dilations = h_33_dilations_0, groups = h_33_groups_0, pad = h_33_pad_0, pad_type = h_33_pad_type_0, strides = h_33_strides_0, weight = layers_16_fc2_weight_to_fp16, x = input_143_cast_fp16)[name = string("h_33_cast_fp16")]; tensor x_183_cast_fp16 = add(x = x_177_cast_fp16, y = h_33_cast_fp16)[name = string("x_183_cast_fp16")]; int32 var_5261 = const()[name = string("op_5261"), val = int32(1)]; tensor mu_69_axes_0 = const()[name = string("mu_69_axes_0"), val = tensor([1])]; bool mu_69_keep_dims_0 = const()[name = string("mu_69_keep_dims_0"), val = bool(true)]; tensor mu_69_cast_fp16 = reduce_mean(axes = mu_69_axes_0, keep_dims = mu_69_keep_dims_0, x = x_183_cast_fp16)[name = string("mu_69_cast_fp16")]; tensor var_5275_cast_fp16 = sub(x = x_183_cast_fp16, y = mu_69_cast_fp16)[name = string("op_5275_cast_fp16")]; fp16 var_5264_promoted_to_fp16 = const()[name = string("op_5264_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_5276_cast_fp16 = pow(x = var_5275_cast_fp16, y = var_5264_promoted_to_fp16)[name = string("op_5276_cast_fp16")]; tensor var_69_axes_0 = const()[name = string("var_69_axes_0"), val = tensor([1])]; bool var_69_keep_dims_0 = const()[name = string("var_69_keep_dims_0"), val = bool(true)]; tensor var_69_cast_fp16 = reduce_mean(axes = var_69_axes_0, keep_dims = var_69_keep_dims_0, x = var_5276_cast_fp16)[name = string("var_69_cast_fp16")]; fp16 var_5280_to_fp16 = const()[name = string("op_5280_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_5281_cast_fp16 = add(x = var_69_cast_fp16, y = var_5280_to_fp16)[name = string("op_5281_cast_fp16")]; fp32 var_5282_epsilon_0 = const()[name = string("op_5282_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_5282_cast_fp16 = rsqrt(epsilon = var_5282_epsilon_0, x = var_5281_cast_fp16)[name = string("op_5282_cast_fp16")]; tensor x_185_cast_fp16 = mul(x = var_5275_cast_fp16, y = var_5282_cast_fp16)[name = string("x_185_cast_fp16")]; tensor input_145_gamma_0_to_fp16 = const()[name = string("input_145_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(452361536)))]; tensor input_145_beta_0_to_fp16 = const()[name = string("input_145_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(452363648)))]; fp16 input_145_epsilon_0_to_fp16 = const()[name = string("input_145_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_145_cast_fp16 = batch_norm(beta = input_145_beta_0_to_fp16, epsilon = input_145_epsilon_0_to_fp16, gamma = input_145_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_185_cast_fp16)[name = string("input_145_cast_fp16")]; string var_5300_pad_type_0 = const()[name = string("op_5300_pad_type_0"), val = string("valid")]; tensor var_5300_strides_0 = const()[name = string("op_5300_strides_0"), val = tensor([1, 1])]; tensor var_5300_pad_0 = const()[name = string("op_5300_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5300_dilations_0 = const()[name = string("op_5300_dilations_0"), val = tensor([1, 1])]; int32 var_5300_groups_0 = const()[name = string("op_5300_groups_0"), val = int32(1)]; tensor var_5302_weight_0_to_fp16 = const()[name = string("op_5302_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(452365760)))]; tensor var_5302_bias_0_to_fp16 = const()[name = string("op_5302_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(454462976)))]; tensor var_5302_cast_fp16 = conv(bias = var_5302_bias_0_to_fp16, dilations = var_5300_dilations_0, groups = var_5300_groups_0, pad = var_5300_pad_0, pad_type = var_5300_pad_type_0, strides = var_5300_strides_0, weight = var_5302_weight_0_to_fp16, x = input_145_cast_fp16)[name = string("op_5302_cast_fp16")]; string var_5309_pad_type_0 = const()[name = string("op_5309_pad_type_0"), val = string("valid")]; tensor var_5309_strides_0 = const()[name = string("op_5309_strides_0"), val = tensor([1, 1])]; tensor var_5309_pad_0 = const()[name = string("op_5309_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5309_dilations_0 = const()[name = string("op_5309_dilations_0"), val = tensor([1, 1])]; int32 var_5309_groups_0 = const()[name = string("op_5309_groups_0"), val = int32(1)]; tensor layers_17_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_17_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(454465088)))]; tensor layers_17_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_17_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(456562304)))]; tensor var_5309_cast_fp16 = conv(bias = layers_17_self_attn_k_proj_bias_to_fp16, dilations = var_5309_dilations_0, groups = var_5309_groups_0, pad = var_5309_pad_0, pad_type = var_5309_pad_type_0, strides = var_5309_strides_0, weight = layers_17_self_attn_k_proj_weight_to_fp16, x = input_145_cast_fp16)[name = string("op_5309_cast_fp16")]; string var_5316_pad_type_0 = const()[name = string("op_5316_pad_type_0"), val = string("valid")]; tensor var_5316_strides_0 = const()[name = string("op_5316_strides_0"), val = tensor([1, 1])]; tensor var_5316_pad_0 = const()[name = string("op_5316_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5316_dilations_0 = const()[name = string("op_5316_dilations_0"), val = tensor([1, 1])]; int32 var_5316_groups_0 = const()[name = string("op_5316_groups_0"), val = int32(1)]; tensor layers_17_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_17_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(456564416)))]; tensor layers_17_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_17_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(458661632)))]; tensor var_5316_cast_fp16 = conv(bias = layers_17_self_attn_v_proj_bias_to_fp16, dilations = var_5316_dilations_0, groups = var_5316_groups_0, pad = var_5316_pad_0, pad_type = var_5316_pad_type_0, strides = var_5316_strides_0, weight = layers_17_self_attn_v_proj_weight_to_fp16, x = input_145_cast_fp16)[name = string("op_5316_cast_fp16")]; tensor tile_51 = const()[name = string("tile_51"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(458663744)))]; int32 var_5317_axis_0 = const()[name = string("op_5317_axis_0"), val = int32(1)]; tensor var_5317_cast_fp16_0, tensor var_5317_cast_fp16_1, tensor var_5317_cast_fp16_2, tensor var_5317_cast_fp16_3, tensor var_5317_cast_fp16_4, tensor var_5317_cast_fp16_5, tensor var_5317_cast_fp16_6, tensor var_5317_cast_fp16_7, tensor var_5317_cast_fp16_8, tensor var_5317_cast_fp16_9, tensor var_5317_cast_fp16_10, tensor var_5317_cast_fp16_11, tensor var_5317_cast_fp16_12, tensor var_5317_cast_fp16_13, tensor var_5317_cast_fp16_14, tensor var_5317_cast_fp16_15 = split(axis = var_5317_axis_0, split_sizes = tile_51, x = var_5302_cast_fp16)[name = string("op_5317_cast_fp16")]; tensor tile_52 = const()[name = string("tile_52"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(458663872)))]; int32 var_5334_axis_0 = const()[name = string("op_5334_axis_0"), val = int32(1)]; tensor var_5334_cast_fp16_0, tensor var_5334_cast_fp16_1, tensor var_5334_cast_fp16_2, tensor var_5334_cast_fp16_3, tensor var_5334_cast_fp16_4, tensor var_5334_cast_fp16_5, tensor var_5334_cast_fp16_6, tensor var_5334_cast_fp16_7, tensor var_5334_cast_fp16_8, tensor var_5334_cast_fp16_9, tensor var_5334_cast_fp16_10, tensor var_5334_cast_fp16_11, tensor var_5334_cast_fp16_12, tensor var_5334_cast_fp16_13, tensor var_5334_cast_fp16_14, tensor var_5334_cast_fp16_15 = split(axis = var_5334_axis_0, split_sizes = tile_52, x = var_5309_cast_fp16)[name = string("op_5334_cast_fp16")]; tensor tile_53 = const()[name = string("tile_53"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(458664000)))]; int32 var_5351_axis_0 = const()[name = string("op_5351_axis_0"), val = int32(1)]; tensor var_5351_cast_fp16_0, tensor var_5351_cast_fp16_1, tensor var_5351_cast_fp16_2, tensor var_5351_cast_fp16_3, tensor var_5351_cast_fp16_4, tensor var_5351_cast_fp16_5, tensor var_5351_cast_fp16_6, tensor var_5351_cast_fp16_7, tensor var_5351_cast_fp16_8, tensor var_5351_cast_fp16_9, tensor var_5351_cast_fp16_10, tensor var_5351_cast_fp16_11, tensor var_5351_cast_fp16_12, tensor var_5351_cast_fp16_13, tensor var_5351_cast_fp16_14, tensor var_5351_cast_fp16_15 = split(axis = var_5351_axis_0, split_sizes = tile_53, x = var_5316_cast_fp16)[name = string("op_5351_cast_fp16")]; tensor transpose_544_perm_0 = const()[name = string("transpose_544_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2724 = const()[name = string("concat_2724"), val = tensor([1, 104, 64])]; tensor transpose_544_cast_fp16 = transpose(perm = transpose_544_perm_0, x = var_5317_cast_fp16_0)[name = string("transpose_3407")]; tensor reshape_816_cast_fp16 = reshape(shape = concat_2724, x = transpose_544_cast_fp16)[name = string("reshape_816_cast_fp16")]; tensor transpose_545_perm_0 = const()[name = string("transpose_545_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2725 = const()[name = string("concat_2725"), val = tensor([1, 64, 104])]; tensor transpose_545_cast_fp16 = transpose(perm = transpose_545_perm_0, x = var_5334_cast_fp16_0)[name = string("transpose_3406")]; tensor reshape_817_cast_fp16 = reshape(shape = concat_2725, x = transpose_545_cast_fp16)[name = string("reshape_817_cast_fp16")]; bool matmul_272_transpose_x_0 = const()[name = string("matmul_272_transpose_x_0"), val = bool(false)]; bool matmul_272_transpose_y_0 = const()[name = string("matmul_272_transpose_y_0"), val = bool(false)]; tensor matmul_272_cast_fp16 = matmul(transpose_x = matmul_272_transpose_x_0, transpose_y = matmul_272_transpose_y_0, x = reshape_816_cast_fp16, y = reshape_817_cast_fp16)[name = string("matmul_272_cast_fp16")]; tensor concat_2729 = const()[name = string("concat_2729"), val = tensor([1, 1, 104, 104])]; tensor reshape_818_cast_fp16 = reshape(shape = concat_2729, x = matmul_272_cast_fp16)[name = string("reshape_818_cast_fp16")]; tensor transpose_2960_perm_0 = const()[name = string("transpose_2960_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2960 = transpose(perm = transpose_2960_perm_0, x = reshape_818_cast_fp16)[name = string("transpose_3405")]; tensor w_1091_cast_fp16 = add(x = transpose_2960, y = transpose_2305)[name = string("w_1091_cast_fp16")]; tensor var_5373_cast_fp16 = softmax(axis = var_5261, x = w_1091_cast_fp16)[name = string("op_5373_cast_fp16")]; string var_5375_equation_0 = const()[name = string("op_5375_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5375_cast_fp16 = einsum(equation = var_5375_equation_0, values = (var_5351_cast_fp16_0, var_5373_cast_fp16))[name = string("op_5375_cast_fp16")]; tensor transpose_546_perm_0 = const()[name = string("transpose_546_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2734 = const()[name = string("concat_2734"), val = tensor([1, 104, 64])]; tensor transpose_546_cast_fp16 = transpose(perm = transpose_546_perm_0, x = var_5317_cast_fp16_1)[name = string("transpose_3404")]; tensor reshape_819_cast_fp16 = reshape(shape = concat_2734, x = transpose_546_cast_fp16)[name = string("reshape_819_cast_fp16")]; tensor transpose_547_perm_0 = const()[name = string("transpose_547_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2735 = const()[name = string("concat_2735"), val = tensor([1, 64, 104])]; tensor transpose_547_cast_fp16 = transpose(perm = transpose_547_perm_0, x = var_5334_cast_fp16_1)[name = string("transpose_3403")]; tensor reshape_820_cast_fp16 = reshape(shape = concat_2735, x = transpose_547_cast_fp16)[name = string("reshape_820_cast_fp16")]; bool matmul_273_transpose_x_0 = const()[name = string("matmul_273_transpose_x_0"), val = bool(false)]; bool matmul_273_transpose_y_0 = const()[name = string("matmul_273_transpose_y_0"), val = bool(false)]; tensor matmul_273_cast_fp16 = matmul(transpose_x = matmul_273_transpose_x_0, transpose_y = matmul_273_transpose_y_0, x = reshape_819_cast_fp16, y = reshape_820_cast_fp16)[name = string("matmul_273_cast_fp16")]; tensor concat_2739 = const()[name = string("concat_2739"), val = tensor([1, 1, 104, 104])]; tensor reshape_821_cast_fp16 = reshape(shape = concat_2739, x = matmul_273_cast_fp16)[name = string("reshape_821_cast_fp16")]; tensor transpose_2961_perm_0 = const()[name = string("transpose_2961_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2961 = transpose(perm = transpose_2961_perm_0, x = reshape_821_cast_fp16)[name = string("transpose_3402")]; tensor w_1095_cast_fp16 = add(x = transpose_2961, y = transpose_2305)[name = string("w_1095_cast_fp16")]; tensor var_5381_cast_fp16 = softmax(axis = var_5261, x = w_1095_cast_fp16)[name = string("op_5381_cast_fp16")]; string var_5383_equation_0 = const()[name = string("op_5383_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5383_cast_fp16 = einsum(equation = var_5383_equation_0, values = (var_5351_cast_fp16_1, var_5381_cast_fp16))[name = string("op_5383_cast_fp16")]; tensor transpose_548_perm_0 = const()[name = string("transpose_548_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2744 = const()[name = string("concat_2744"), val = tensor([1, 104, 64])]; tensor transpose_548_cast_fp16 = transpose(perm = transpose_548_perm_0, x = var_5317_cast_fp16_2)[name = string("transpose_3401")]; tensor reshape_822_cast_fp16 = reshape(shape = concat_2744, x = transpose_548_cast_fp16)[name = string("reshape_822_cast_fp16")]; tensor transpose_549_perm_0 = const()[name = string("transpose_549_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2745 = const()[name = string("concat_2745"), val = tensor([1, 64, 104])]; tensor transpose_549_cast_fp16 = transpose(perm = transpose_549_perm_0, x = var_5334_cast_fp16_2)[name = string("transpose_3400")]; tensor reshape_823_cast_fp16 = reshape(shape = concat_2745, x = transpose_549_cast_fp16)[name = string("reshape_823_cast_fp16")]; bool matmul_274_transpose_x_0 = const()[name = string("matmul_274_transpose_x_0"), val = bool(false)]; bool matmul_274_transpose_y_0 = const()[name = string("matmul_274_transpose_y_0"), val = bool(false)]; tensor matmul_274_cast_fp16 = matmul(transpose_x = matmul_274_transpose_x_0, transpose_y = matmul_274_transpose_y_0, x = reshape_822_cast_fp16, y = reshape_823_cast_fp16)[name = string("matmul_274_cast_fp16")]; tensor concat_2749 = const()[name = string("concat_2749"), val = tensor([1, 1, 104, 104])]; tensor reshape_824_cast_fp16 = reshape(shape = concat_2749, x = matmul_274_cast_fp16)[name = string("reshape_824_cast_fp16")]; tensor transpose_2962_perm_0 = const()[name = string("transpose_2962_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2962 = transpose(perm = transpose_2962_perm_0, x = reshape_824_cast_fp16)[name = string("transpose_3399")]; tensor w_1099_cast_fp16 = add(x = transpose_2962, y = transpose_2305)[name = string("w_1099_cast_fp16")]; tensor var_5389_cast_fp16 = softmax(axis = var_5261, x = w_1099_cast_fp16)[name = string("op_5389_cast_fp16")]; string var_5391_equation_0 = const()[name = string("op_5391_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5391_cast_fp16 = einsum(equation = var_5391_equation_0, values = (var_5351_cast_fp16_2, var_5389_cast_fp16))[name = string("op_5391_cast_fp16")]; tensor transpose_550_perm_0 = const()[name = string("transpose_550_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2754 = const()[name = string("concat_2754"), val = tensor([1, 104, 64])]; tensor transpose_550_cast_fp16 = transpose(perm = transpose_550_perm_0, x = var_5317_cast_fp16_3)[name = string("transpose_3398")]; tensor reshape_825_cast_fp16 = reshape(shape = concat_2754, x = transpose_550_cast_fp16)[name = string("reshape_825_cast_fp16")]; tensor transpose_551_perm_0 = const()[name = string("transpose_551_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2755 = const()[name = string("concat_2755"), val = tensor([1, 64, 104])]; tensor transpose_551_cast_fp16 = transpose(perm = transpose_551_perm_0, x = var_5334_cast_fp16_3)[name = string("transpose_3397")]; tensor reshape_826_cast_fp16 = reshape(shape = concat_2755, x = transpose_551_cast_fp16)[name = string("reshape_826_cast_fp16")]; bool matmul_275_transpose_x_0 = const()[name = string("matmul_275_transpose_x_0"), val = bool(false)]; bool matmul_275_transpose_y_0 = const()[name = string("matmul_275_transpose_y_0"), val = bool(false)]; tensor matmul_275_cast_fp16 = matmul(transpose_x = matmul_275_transpose_x_0, transpose_y = matmul_275_transpose_y_0, x = reshape_825_cast_fp16, y = reshape_826_cast_fp16)[name = string("matmul_275_cast_fp16")]; tensor concat_2759 = const()[name = string("concat_2759"), val = tensor([1, 1, 104, 104])]; tensor reshape_827_cast_fp16 = reshape(shape = concat_2759, x = matmul_275_cast_fp16)[name = string("reshape_827_cast_fp16")]; tensor transpose_2963_perm_0 = const()[name = string("transpose_2963_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2963 = transpose(perm = transpose_2963_perm_0, x = reshape_827_cast_fp16)[name = string("transpose_3396")]; tensor w_1103_cast_fp16 = add(x = transpose_2963, y = transpose_2305)[name = string("w_1103_cast_fp16")]; tensor var_5397_cast_fp16 = softmax(axis = var_5261, x = w_1103_cast_fp16)[name = string("op_5397_cast_fp16")]; string var_5399_equation_0 = const()[name = string("op_5399_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5399_cast_fp16 = einsum(equation = var_5399_equation_0, values = (var_5351_cast_fp16_3, var_5397_cast_fp16))[name = string("op_5399_cast_fp16")]; tensor transpose_552_perm_0 = const()[name = string("transpose_552_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2764 = const()[name = string("concat_2764"), val = tensor([1, 104, 64])]; tensor transpose_552_cast_fp16 = transpose(perm = transpose_552_perm_0, x = var_5317_cast_fp16_4)[name = string("transpose_3395")]; tensor reshape_828_cast_fp16 = reshape(shape = concat_2764, x = transpose_552_cast_fp16)[name = string("reshape_828_cast_fp16")]; tensor transpose_553_perm_0 = const()[name = string("transpose_553_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2765 = const()[name = string("concat_2765"), val = tensor([1, 64, 104])]; tensor transpose_553_cast_fp16 = transpose(perm = transpose_553_perm_0, x = var_5334_cast_fp16_4)[name = string("transpose_3394")]; tensor reshape_829_cast_fp16 = reshape(shape = concat_2765, x = transpose_553_cast_fp16)[name = string("reshape_829_cast_fp16")]; bool matmul_276_transpose_x_0 = const()[name = string("matmul_276_transpose_x_0"), val = bool(false)]; bool matmul_276_transpose_y_0 = const()[name = string("matmul_276_transpose_y_0"), val = bool(false)]; tensor matmul_276_cast_fp16 = matmul(transpose_x = matmul_276_transpose_x_0, transpose_y = matmul_276_transpose_y_0, x = reshape_828_cast_fp16, y = reshape_829_cast_fp16)[name = string("matmul_276_cast_fp16")]; tensor concat_2769 = const()[name = string("concat_2769"), val = tensor([1, 1, 104, 104])]; tensor reshape_830_cast_fp16 = reshape(shape = concat_2769, x = matmul_276_cast_fp16)[name = string("reshape_830_cast_fp16")]; tensor transpose_2964_perm_0 = const()[name = string("transpose_2964_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2964 = transpose(perm = transpose_2964_perm_0, x = reshape_830_cast_fp16)[name = string("transpose_3393")]; tensor w_1107_cast_fp16 = add(x = transpose_2964, y = transpose_2305)[name = string("w_1107_cast_fp16")]; tensor var_5405_cast_fp16 = softmax(axis = var_5261, x = w_1107_cast_fp16)[name = string("op_5405_cast_fp16")]; string var_5407_equation_0 = const()[name = string("op_5407_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5407_cast_fp16 = einsum(equation = var_5407_equation_0, values = (var_5351_cast_fp16_4, var_5405_cast_fp16))[name = string("op_5407_cast_fp16")]; tensor transpose_554_perm_0 = const()[name = string("transpose_554_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2774 = const()[name = string("concat_2774"), val = tensor([1, 104, 64])]; tensor transpose_554_cast_fp16 = transpose(perm = transpose_554_perm_0, x = var_5317_cast_fp16_5)[name = string("transpose_3392")]; tensor reshape_831_cast_fp16 = reshape(shape = concat_2774, x = transpose_554_cast_fp16)[name = string("reshape_831_cast_fp16")]; tensor transpose_555_perm_0 = const()[name = string("transpose_555_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2775 = const()[name = string("concat_2775"), val = tensor([1, 64, 104])]; tensor transpose_555_cast_fp16 = transpose(perm = transpose_555_perm_0, x = var_5334_cast_fp16_5)[name = string("transpose_3391")]; tensor reshape_832_cast_fp16 = reshape(shape = concat_2775, x = transpose_555_cast_fp16)[name = string("reshape_832_cast_fp16")]; bool matmul_277_transpose_x_0 = const()[name = string("matmul_277_transpose_x_0"), val = bool(false)]; bool matmul_277_transpose_y_0 = const()[name = string("matmul_277_transpose_y_0"), val = bool(false)]; tensor matmul_277_cast_fp16 = matmul(transpose_x = matmul_277_transpose_x_0, transpose_y = matmul_277_transpose_y_0, x = reshape_831_cast_fp16, y = reshape_832_cast_fp16)[name = string("matmul_277_cast_fp16")]; tensor concat_2779 = const()[name = string("concat_2779"), val = tensor([1, 1, 104, 104])]; tensor reshape_833_cast_fp16 = reshape(shape = concat_2779, x = matmul_277_cast_fp16)[name = string("reshape_833_cast_fp16")]; tensor transpose_2965_perm_0 = const()[name = string("transpose_2965_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2965 = transpose(perm = transpose_2965_perm_0, x = reshape_833_cast_fp16)[name = string("transpose_3390")]; tensor w_1111_cast_fp16 = add(x = transpose_2965, y = transpose_2305)[name = string("w_1111_cast_fp16")]; tensor var_5413_cast_fp16 = softmax(axis = var_5261, x = w_1111_cast_fp16)[name = string("op_5413_cast_fp16")]; string var_5415_equation_0 = const()[name = string("op_5415_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5415_cast_fp16 = einsum(equation = var_5415_equation_0, values = (var_5351_cast_fp16_5, var_5413_cast_fp16))[name = string("op_5415_cast_fp16")]; tensor transpose_556_perm_0 = const()[name = string("transpose_556_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2784 = const()[name = string("concat_2784"), val = tensor([1, 104, 64])]; tensor transpose_556_cast_fp16 = transpose(perm = transpose_556_perm_0, x = var_5317_cast_fp16_6)[name = string("transpose_3389")]; tensor reshape_834_cast_fp16 = reshape(shape = concat_2784, x = transpose_556_cast_fp16)[name = string("reshape_834_cast_fp16")]; tensor transpose_557_perm_0 = const()[name = string("transpose_557_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2785 = const()[name = string("concat_2785"), val = tensor([1, 64, 104])]; tensor transpose_557_cast_fp16 = transpose(perm = transpose_557_perm_0, x = var_5334_cast_fp16_6)[name = string("transpose_3388")]; tensor reshape_835_cast_fp16 = reshape(shape = concat_2785, x = transpose_557_cast_fp16)[name = string("reshape_835_cast_fp16")]; bool matmul_278_transpose_x_0 = const()[name = string("matmul_278_transpose_x_0"), val = bool(false)]; bool matmul_278_transpose_y_0 = const()[name = string("matmul_278_transpose_y_0"), val = bool(false)]; tensor matmul_278_cast_fp16 = matmul(transpose_x = matmul_278_transpose_x_0, transpose_y = matmul_278_transpose_y_0, x = reshape_834_cast_fp16, y = reshape_835_cast_fp16)[name = string("matmul_278_cast_fp16")]; tensor concat_2789 = const()[name = string("concat_2789"), val = tensor([1, 1, 104, 104])]; tensor reshape_836_cast_fp16 = reshape(shape = concat_2789, x = matmul_278_cast_fp16)[name = string("reshape_836_cast_fp16")]; tensor transpose_2966_perm_0 = const()[name = string("transpose_2966_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2966 = transpose(perm = transpose_2966_perm_0, x = reshape_836_cast_fp16)[name = string("transpose_3387")]; tensor w_1115_cast_fp16 = add(x = transpose_2966, y = transpose_2305)[name = string("w_1115_cast_fp16")]; tensor var_5421_cast_fp16 = softmax(axis = var_5261, x = w_1115_cast_fp16)[name = string("op_5421_cast_fp16")]; string var_5423_equation_0 = const()[name = string("op_5423_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5423_cast_fp16 = einsum(equation = var_5423_equation_0, values = (var_5351_cast_fp16_6, var_5421_cast_fp16))[name = string("op_5423_cast_fp16")]; tensor transpose_558_perm_0 = const()[name = string("transpose_558_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2794 = const()[name = string("concat_2794"), val = tensor([1, 104, 64])]; tensor transpose_558_cast_fp16 = transpose(perm = transpose_558_perm_0, x = var_5317_cast_fp16_7)[name = string("transpose_3386")]; tensor reshape_837_cast_fp16 = reshape(shape = concat_2794, x = transpose_558_cast_fp16)[name = string("reshape_837_cast_fp16")]; tensor transpose_559_perm_0 = const()[name = string("transpose_559_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2795 = const()[name = string("concat_2795"), val = tensor([1, 64, 104])]; tensor transpose_559_cast_fp16 = transpose(perm = transpose_559_perm_0, x = var_5334_cast_fp16_7)[name = string("transpose_3385")]; tensor reshape_838_cast_fp16 = reshape(shape = concat_2795, x = transpose_559_cast_fp16)[name = string("reshape_838_cast_fp16")]; bool matmul_279_transpose_x_0 = const()[name = string("matmul_279_transpose_x_0"), val = bool(false)]; bool matmul_279_transpose_y_0 = const()[name = string("matmul_279_transpose_y_0"), val = bool(false)]; tensor matmul_279_cast_fp16 = matmul(transpose_x = matmul_279_transpose_x_0, transpose_y = matmul_279_transpose_y_0, x = reshape_837_cast_fp16, y = reshape_838_cast_fp16)[name = string("matmul_279_cast_fp16")]; tensor concat_2799 = const()[name = string("concat_2799"), val = tensor([1, 1, 104, 104])]; tensor reshape_839_cast_fp16 = reshape(shape = concat_2799, x = matmul_279_cast_fp16)[name = string("reshape_839_cast_fp16")]; tensor transpose_2967_perm_0 = const()[name = string("transpose_2967_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2967 = transpose(perm = transpose_2967_perm_0, x = reshape_839_cast_fp16)[name = string("transpose_3384")]; tensor w_1119_cast_fp16 = add(x = transpose_2967, y = transpose_2305)[name = string("w_1119_cast_fp16")]; tensor var_5429_cast_fp16 = softmax(axis = var_5261, x = w_1119_cast_fp16)[name = string("op_5429_cast_fp16")]; string var_5431_equation_0 = const()[name = string("op_5431_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5431_cast_fp16 = einsum(equation = var_5431_equation_0, values = (var_5351_cast_fp16_7, var_5429_cast_fp16))[name = string("op_5431_cast_fp16")]; tensor transpose_560_perm_0 = const()[name = string("transpose_560_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2804 = const()[name = string("concat_2804"), val = tensor([1, 104, 64])]; tensor transpose_560_cast_fp16 = transpose(perm = transpose_560_perm_0, x = var_5317_cast_fp16_8)[name = string("transpose_3383")]; tensor reshape_840_cast_fp16 = reshape(shape = concat_2804, x = transpose_560_cast_fp16)[name = string("reshape_840_cast_fp16")]; tensor transpose_561_perm_0 = const()[name = string("transpose_561_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2805 = const()[name = string("concat_2805"), val = tensor([1, 64, 104])]; tensor transpose_561_cast_fp16 = transpose(perm = transpose_561_perm_0, x = var_5334_cast_fp16_8)[name = string("transpose_3382")]; tensor reshape_841_cast_fp16 = reshape(shape = concat_2805, x = transpose_561_cast_fp16)[name = string("reshape_841_cast_fp16")]; bool matmul_280_transpose_x_0 = const()[name = string("matmul_280_transpose_x_0"), val = bool(false)]; bool matmul_280_transpose_y_0 = const()[name = string("matmul_280_transpose_y_0"), val = bool(false)]; tensor matmul_280_cast_fp16 = matmul(transpose_x = matmul_280_transpose_x_0, transpose_y = matmul_280_transpose_y_0, x = reshape_840_cast_fp16, y = reshape_841_cast_fp16)[name = string("matmul_280_cast_fp16")]; tensor concat_2809 = const()[name = string("concat_2809"), val = tensor([1, 1, 104, 104])]; tensor reshape_842_cast_fp16 = reshape(shape = concat_2809, x = matmul_280_cast_fp16)[name = string("reshape_842_cast_fp16")]; tensor transpose_2968_perm_0 = const()[name = string("transpose_2968_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2968 = transpose(perm = transpose_2968_perm_0, x = reshape_842_cast_fp16)[name = string("transpose_3381")]; tensor w_1123_cast_fp16 = add(x = transpose_2968, y = transpose_2305)[name = string("w_1123_cast_fp16")]; tensor var_5437_cast_fp16 = softmax(axis = var_5261, x = w_1123_cast_fp16)[name = string("op_5437_cast_fp16")]; string var_5439_equation_0 = const()[name = string("op_5439_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5439_cast_fp16 = einsum(equation = var_5439_equation_0, values = (var_5351_cast_fp16_8, var_5437_cast_fp16))[name = string("op_5439_cast_fp16")]; tensor transpose_562_perm_0 = const()[name = string("transpose_562_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2814 = const()[name = string("concat_2814"), val = tensor([1, 104, 64])]; tensor transpose_562_cast_fp16 = transpose(perm = transpose_562_perm_0, x = var_5317_cast_fp16_9)[name = string("transpose_3380")]; tensor reshape_843_cast_fp16 = reshape(shape = concat_2814, x = transpose_562_cast_fp16)[name = string("reshape_843_cast_fp16")]; tensor transpose_563_perm_0 = const()[name = string("transpose_563_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2815 = const()[name = string("concat_2815"), val = tensor([1, 64, 104])]; tensor transpose_563_cast_fp16 = transpose(perm = transpose_563_perm_0, x = var_5334_cast_fp16_9)[name = string("transpose_3379")]; tensor reshape_844_cast_fp16 = reshape(shape = concat_2815, x = transpose_563_cast_fp16)[name = string("reshape_844_cast_fp16")]; bool matmul_281_transpose_x_0 = const()[name = string("matmul_281_transpose_x_0"), val = bool(false)]; bool matmul_281_transpose_y_0 = const()[name = string("matmul_281_transpose_y_0"), val = bool(false)]; tensor matmul_281_cast_fp16 = matmul(transpose_x = matmul_281_transpose_x_0, transpose_y = matmul_281_transpose_y_0, x = reshape_843_cast_fp16, y = reshape_844_cast_fp16)[name = string("matmul_281_cast_fp16")]; tensor concat_2819 = const()[name = string("concat_2819"), val = tensor([1, 1, 104, 104])]; tensor reshape_845_cast_fp16 = reshape(shape = concat_2819, x = matmul_281_cast_fp16)[name = string("reshape_845_cast_fp16")]; tensor transpose_2969_perm_0 = const()[name = string("transpose_2969_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2969 = transpose(perm = transpose_2969_perm_0, x = reshape_845_cast_fp16)[name = string("transpose_3378")]; tensor w_1127_cast_fp16 = add(x = transpose_2969, y = transpose_2305)[name = string("w_1127_cast_fp16")]; tensor var_5445_cast_fp16 = softmax(axis = var_5261, x = w_1127_cast_fp16)[name = string("op_5445_cast_fp16")]; string var_5447_equation_0 = const()[name = string("op_5447_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5447_cast_fp16 = einsum(equation = var_5447_equation_0, values = (var_5351_cast_fp16_9, var_5445_cast_fp16))[name = string("op_5447_cast_fp16")]; tensor transpose_564_perm_0 = const()[name = string("transpose_564_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2824 = const()[name = string("concat_2824"), val = tensor([1, 104, 64])]; tensor transpose_564_cast_fp16 = transpose(perm = transpose_564_perm_0, x = var_5317_cast_fp16_10)[name = string("transpose_3377")]; tensor reshape_846_cast_fp16 = reshape(shape = concat_2824, x = transpose_564_cast_fp16)[name = string("reshape_846_cast_fp16")]; tensor transpose_565_perm_0 = const()[name = string("transpose_565_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2825 = const()[name = string("concat_2825"), val = tensor([1, 64, 104])]; tensor transpose_565_cast_fp16 = transpose(perm = transpose_565_perm_0, x = var_5334_cast_fp16_10)[name = string("transpose_3376")]; tensor reshape_847_cast_fp16 = reshape(shape = concat_2825, x = transpose_565_cast_fp16)[name = string("reshape_847_cast_fp16")]; bool matmul_282_transpose_x_0 = const()[name = string("matmul_282_transpose_x_0"), val = bool(false)]; bool matmul_282_transpose_y_0 = const()[name = string("matmul_282_transpose_y_0"), val = bool(false)]; tensor matmul_282_cast_fp16 = matmul(transpose_x = matmul_282_transpose_x_0, transpose_y = matmul_282_transpose_y_0, x = reshape_846_cast_fp16, y = reshape_847_cast_fp16)[name = string("matmul_282_cast_fp16")]; tensor concat_2829 = const()[name = string("concat_2829"), val = tensor([1, 1, 104, 104])]; tensor reshape_848_cast_fp16 = reshape(shape = concat_2829, x = matmul_282_cast_fp16)[name = string("reshape_848_cast_fp16")]; tensor transpose_2970_perm_0 = const()[name = string("transpose_2970_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2970 = transpose(perm = transpose_2970_perm_0, x = reshape_848_cast_fp16)[name = string("transpose_3375")]; tensor w_1131_cast_fp16 = add(x = transpose_2970, y = transpose_2305)[name = string("w_1131_cast_fp16")]; tensor var_5453_cast_fp16 = softmax(axis = var_5261, x = w_1131_cast_fp16)[name = string("op_5453_cast_fp16")]; string var_5455_equation_0 = const()[name = string("op_5455_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5455_cast_fp16 = einsum(equation = var_5455_equation_0, values = (var_5351_cast_fp16_10, var_5453_cast_fp16))[name = string("op_5455_cast_fp16")]; tensor transpose_566_perm_0 = const()[name = string("transpose_566_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2834 = const()[name = string("concat_2834"), val = tensor([1, 104, 64])]; tensor transpose_566_cast_fp16 = transpose(perm = transpose_566_perm_0, x = var_5317_cast_fp16_11)[name = string("transpose_3374")]; tensor reshape_849_cast_fp16 = reshape(shape = concat_2834, x = transpose_566_cast_fp16)[name = string("reshape_849_cast_fp16")]; tensor transpose_567_perm_0 = const()[name = string("transpose_567_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2835 = const()[name = string("concat_2835"), val = tensor([1, 64, 104])]; tensor transpose_567_cast_fp16 = transpose(perm = transpose_567_perm_0, x = var_5334_cast_fp16_11)[name = string("transpose_3373")]; tensor reshape_850_cast_fp16 = reshape(shape = concat_2835, x = transpose_567_cast_fp16)[name = string("reshape_850_cast_fp16")]; bool matmul_283_transpose_x_0 = const()[name = string("matmul_283_transpose_x_0"), val = bool(false)]; bool matmul_283_transpose_y_0 = const()[name = string("matmul_283_transpose_y_0"), val = bool(false)]; tensor matmul_283_cast_fp16 = matmul(transpose_x = matmul_283_transpose_x_0, transpose_y = matmul_283_transpose_y_0, x = reshape_849_cast_fp16, y = reshape_850_cast_fp16)[name = string("matmul_283_cast_fp16")]; tensor concat_2839 = const()[name = string("concat_2839"), val = tensor([1, 1, 104, 104])]; tensor reshape_851_cast_fp16 = reshape(shape = concat_2839, x = matmul_283_cast_fp16)[name = string("reshape_851_cast_fp16")]; tensor transpose_2971_perm_0 = const()[name = string("transpose_2971_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2971 = transpose(perm = transpose_2971_perm_0, x = reshape_851_cast_fp16)[name = string("transpose_3372")]; tensor w_1135_cast_fp16 = add(x = transpose_2971, y = transpose_2305)[name = string("w_1135_cast_fp16")]; tensor var_5461_cast_fp16 = softmax(axis = var_5261, x = w_1135_cast_fp16)[name = string("op_5461_cast_fp16")]; string var_5463_equation_0 = const()[name = string("op_5463_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5463_cast_fp16 = einsum(equation = var_5463_equation_0, values = (var_5351_cast_fp16_11, var_5461_cast_fp16))[name = string("op_5463_cast_fp16")]; tensor transpose_568_perm_0 = const()[name = string("transpose_568_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2844 = const()[name = string("concat_2844"), val = tensor([1, 104, 64])]; tensor transpose_568_cast_fp16 = transpose(perm = transpose_568_perm_0, x = var_5317_cast_fp16_12)[name = string("transpose_3371")]; tensor reshape_852_cast_fp16 = reshape(shape = concat_2844, x = transpose_568_cast_fp16)[name = string("reshape_852_cast_fp16")]; tensor transpose_569_perm_0 = const()[name = string("transpose_569_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2845 = const()[name = string("concat_2845"), val = tensor([1, 64, 104])]; tensor transpose_569_cast_fp16 = transpose(perm = transpose_569_perm_0, x = var_5334_cast_fp16_12)[name = string("transpose_3370")]; tensor reshape_853_cast_fp16 = reshape(shape = concat_2845, x = transpose_569_cast_fp16)[name = string("reshape_853_cast_fp16")]; bool matmul_284_transpose_x_0 = const()[name = string("matmul_284_transpose_x_0"), val = bool(false)]; bool matmul_284_transpose_y_0 = const()[name = string("matmul_284_transpose_y_0"), val = bool(false)]; tensor matmul_284_cast_fp16 = matmul(transpose_x = matmul_284_transpose_x_0, transpose_y = matmul_284_transpose_y_0, x = reshape_852_cast_fp16, y = reshape_853_cast_fp16)[name = string("matmul_284_cast_fp16")]; tensor concat_2849 = const()[name = string("concat_2849"), val = tensor([1, 1, 104, 104])]; tensor reshape_854_cast_fp16 = reshape(shape = concat_2849, x = matmul_284_cast_fp16)[name = string("reshape_854_cast_fp16")]; tensor transpose_2972_perm_0 = const()[name = string("transpose_2972_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2972 = transpose(perm = transpose_2972_perm_0, x = reshape_854_cast_fp16)[name = string("transpose_3369")]; tensor w_1139_cast_fp16 = add(x = transpose_2972, y = transpose_2305)[name = string("w_1139_cast_fp16")]; tensor var_5469_cast_fp16 = softmax(axis = var_5261, x = w_1139_cast_fp16)[name = string("op_5469_cast_fp16")]; string var_5471_equation_0 = const()[name = string("op_5471_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5471_cast_fp16 = einsum(equation = var_5471_equation_0, values = (var_5351_cast_fp16_12, var_5469_cast_fp16))[name = string("op_5471_cast_fp16")]; tensor transpose_570_perm_0 = const()[name = string("transpose_570_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2854 = const()[name = string("concat_2854"), val = tensor([1, 104, 64])]; tensor transpose_570_cast_fp16 = transpose(perm = transpose_570_perm_0, x = var_5317_cast_fp16_13)[name = string("transpose_3368")]; tensor reshape_855_cast_fp16 = reshape(shape = concat_2854, x = transpose_570_cast_fp16)[name = string("reshape_855_cast_fp16")]; tensor transpose_571_perm_0 = const()[name = string("transpose_571_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2855 = const()[name = string("concat_2855"), val = tensor([1, 64, 104])]; tensor transpose_571_cast_fp16 = transpose(perm = transpose_571_perm_0, x = var_5334_cast_fp16_13)[name = string("transpose_3367")]; tensor reshape_856_cast_fp16 = reshape(shape = concat_2855, x = transpose_571_cast_fp16)[name = string("reshape_856_cast_fp16")]; bool matmul_285_transpose_x_0 = const()[name = string("matmul_285_transpose_x_0"), val = bool(false)]; bool matmul_285_transpose_y_0 = const()[name = string("matmul_285_transpose_y_0"), val = bool(false)]; tensor matmul_285_cast_fp16 = matmul(transpose_x = matmul_285_transpose_x_0, transpose_y = matmul_285_transpose_y_0, x = reshape_855_cast_fp16, y = reshape_856_cast_fp16)[name = string("matmul_285_cast_fp16")]; tensor concat_2859 = const()[name = string("concat_2859"), val = tensor([1, 1, 104, 104])]; tensor reshape_857_cast_fp16 = reshape(shape = concat_2859, x = matmul_285_cast_fp16)[name = string("reshape_857_cast_fp16")]; tensor transpose_2973_perm_0 = const()[name = string("transpose_2973_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2973 = transpose(perm = transpose_2973_perm_0, x = reshape_857_cast_fp16)[name = string("transpose_3366")]; tensor w_1143_cast_fp16 = add(x = transpose_2973, y = transpose_2305)[name = string("w_1143_cast_fp16")]; tensor var_5477_cast_fp16 = softmax(axis = var_5261, x = w_1143_cast_fp16)[name = string("op_5477_cast_fp16")]; string var_5479_equation_0 = const()[name = string("op_5479_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5479_cast_fp16 = einsum(equation = var_5479_equation_0, values = (var_5351_cast_fp16_13, var_5477_cast_fp16))[name = string("op_5479_cast_fp16")]; tensor transpose_572_perm_0 = const()[name = string("transpose_572_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2864 = const()[name = string("concat_2864"), val = tensor([1, 104, 64])]; tensor transpose_572_cast_fp16 = transpose(perm = transpose_572_perm_0, x = var_5317_cast_fp16_14)[name = string("transpose_3365")]; tensor reshape_858_cast_fp16 = reshape(shape = concat_2864, x = transpose_572_cast_fp16)[name = string("reshape_858_cast_fp16")]; tensor transpose_573_perm_0 = const()[name = string("transpose_573_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2865 = const()[name = string("concat_2865"), val = tensor([1, 64, 104])]; tensor transpose_573_cast_fp16 = transpose(perm = transpose_573_perm_0, x = var_5334_cast_fp16_14)[name = string("transpose_3364")]; tensor reshape_859_cast_fp16 = reshape(shape = concat_2865, x = transpose_573_cast_fp16)[name = string("reshape_859_cast_fp16")]; bool matmul_286_transpose_x_0 = const()[name = string("matmul_286_transpose_x_0"), val = bool(false)]; bool matmul_286_transpose_y_0 = const()[name = string("matmul_286_transpose_y_0"), val = bool(false)]; tensor matmul_286_cast_fp16 = matmul(transpose_x = matmul_286_transpose_x_0, transpose_y = matmul_286_transpose_y_0, x = reshape_858_cast_fp16, y = reshape_859_cast_fp16)[name = string("matmul_286_cast_fp16")]; tensor concat_2869 = const()[name = string("concat_2869"), val = tensor([1, 1, 104, 104])]; tensor reshape_860_cast_fp16 = reshape(shape = concat_2869, x = matmul_286_cast_fp16)[name = string("reshape_860_cast_fp16")]; tensor transpose_2974_perm_0 = const()[name = string("transpose_2974_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2974 = transpose(perm = transpose_2974_perm_0, x = reshape_860_cast_fp16)[name = string("transpose_3363")]; tensor w_1147_cast_fp16 = add(x = transpose_2974, y = transpose_2305)[name = string("w_1147_cast_fp16")]; tensor var_5485_cast_fp16 = softmax(axis = var_5261, x = w_1147_cast_fp16)[name = string("op_5485_cast_fp16")]; string var_5487_equation_0 = const()[name = string("op_5487_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5487_cast_fp16 = einsum(equation = var_5487_equation_0, values = (var_5351_cast_fp16_14, var_5485_cast_fp16))[name = string("op_5487_cast_fp16")]; tensor transpose_574_perm_0 = const()[name = string("transpose_574_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2874 = const()[name = string("concat_2874"), val = tensor([1, 104, 64])]; tensor transpose_574_cast_fp16 = transpose(perm = transpose_574_perm_0, x = var_5317_cast_fp16_15)[name = string("transpose_3362")]; tensor reshape_861_cast_fp16 = reshape(shape = concat_2874, x = transpose_574_cast_fp16)[name = string("reshape_861_cast_fp16")]; tensor transpose_575_perm_0 = const()[name = string("transpose_575_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2875 = const()[name = string("concat_2875"), val = tensor([1, 64, 104])]; tensor transpose_575_cast_fp16 = transpose(perm = transpose_575_perm_0, x = var_5334_cast_fp16_15)[name = string("transpose_3361")]; tensor reshape_862_cast_fp16 = reshape(shape = concat_2875, x = transpose_575_cast_fp16)[name = string("reshape_862_cast_fp16")]; bool matmul_287_transpose_x_0 = const()[name = string("matmul_287_transpose_x_0"), val = bool(false)]; bool matmul_287_transpose_y_0 = const()[name = string("matmul_287_transpose_y_0"), val = bool(false)]; tensor matmul_287_cast_fp16 = matmul(transpose_x = matmul_287_transpose_x_0, transpose_y = matmul_287_transpose_y_0, x = reshape_861_cast_fp16, y = reshape_862_cast_fp16)[name = string("matmul_287_cast_fp16")]; tensor concat_2879 = const()[name = string("concat_2879"), val = tensor([1, 1, 104, 104])]; tensor reshape_863_cast_fp16 = reshape(shape = concat_2879, x = matmul_287_cast_fp16)[name = string("reshape_863_cast_fp16")]; tensor transpose_2975_perm_0 = const()[name = string("transpose_2975_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2975 = transpose(perm = transpose_2975_perm_0, x = reshape_863_cast_fp16)[name = string("transpose_3360")]; tensor w_1151_cast_fp16 = add(x = transpose_2975, y = transpose_2305)[name = string("w_1151_cast_fp16")]; tensor var_5493_cast_fp16 = softmax(axis = var_5261, x = w_1151_cast_fp16)[name = string("op_5493_cast_fp16")]; string var_5495_equation_0 = const()[name = string("op_5495_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5495_cast_fp16 = einsum(equation = var_5495_equation_0, values = (var_5351_cast_fp16_15, var_5493_cast_fp16))[name = string("op_5495_cast_fp16")]; bool input_147_interleave_0 = const()[name = string("input_147_interleave_0"), val = bool(false)]; tensor input_147_cast_fp16 = concat(axis = var_5261, interleave = input_147_interleave_0, values = (var_5375_cast_fp16, var_5383_cast_fp16, var_5391_cast_fp16, var_5399_cast_fp16, var_5407_cast_fp16, var_5415_cast_fp16, var_5423_cast_fp16, var_5431_cast_fp16, var_5439_cast_fp16, var_5447_cast_fp16, var_5455_cast_fp16, var_5463_cast_fp16, var_5471_cast_fp16, var_5479_cast_fp16, var_5487_cast_fp16, var_5495_cast_fp16))[name = string("input_147_cast_fp16")]; string var_5504_pad_type_0 = const()[name = string("op_5504_pad_type_0"), val = string("valid")]; tensor var_5504_strides_0 = const()[name = string("op_5504_strides_0"), val = tensor([1, 1])]; tensor var_5504_pad_0 = const()[name = string("op_5504_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5504_dilations_0 = const()[name = string("op_5504_dilations_0"), val = tensor([1, 1])]; int32 var_5504_groups_0 = const()[name = string("op_5504_groups_0"), val = int32(1)]; tensor layers_17_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_17_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(458664128)))]; tensor layers_17_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_17_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(460761344)))]; tensor var_5504_cast_fp16 = conv(bias = layers_17_self_attn_out_proj_bias_to_fp16, dilations = var_5504_dilations_0, groups = var_5504_groups_0, pad = var_5504_pad_0, pad_type = var_5504_pad_type_0, strides = var_5504_strides_0, weight = layers_17_self_attn_out_proj_weight_to_fp16, x = input_147_cast_fp16)[name = string("op_5504_cast_fp16")]; tensor x_187_cast_fp16 = add(x = x_183_cast_fp16, y = var_5504_cast_fp16)[name = string("x_187_cast_fp16")]; tensor mu_71_axes_0 = const()[name = string("mu_71_axes_0"), val = tensor([1])]; bool mu_71_keep_dims_0 = const()[name = string("mu_71_keep_dims_0"), val = bool(true)]; tensor mu_71_cast_fp16 = reduce_mean(axes = mu_71_axes_0, keep_dims = mu_71_keep_dims_0, x = x_187_cast_fp16)[name = string("mu_71_cast_fp16")]; tensor var_5510_cast_fp16 = sub(x = x_187_cast_fp16, y = mu_71_cast_fp16)[name = string("op_5510_cast_fp16")]; fp16 var_5264_promoted_1_to_fp16 = const()[name = string("op_5264_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_5511_cast_fp16 = pow(x = var_5510_cast_fp16, y = var_5264_promoted_1_to_fp16)[name = string("op_5511_cast_fp16")]; tensor var_71_axes_0 = const()[name = string("var_71_axes_0"), val = tensor([1])]; bool var_71_keep_dims_0 = const()[name = string("var_71_keep_dims_0"), val = bool(true)]; tensor var_71_cast_fp16 = reduce_mean(axes = var_71_axes_0, keep_dims = var_71_keep_dims_0, x = var_5511_cast_fp16)[name = string("var_71_cast_fp16")]; fp16 var_5515_to_fp16 = const()[name = string("op_5515_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_5516_cast_fp16 = add(x = var_71_cast_fp16, y = var_5515_to_fp16)[name = string("op_5516_cast_fp16")]; fp32 var_5517_epsilon_0 = const()[name = string("op_5517_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_5517_cast_fp16 = rsqrt(epsilon = var_5517_epsilon_0, x = var_5516_cast_fp16)[name = string("op_5517_cast_fp16")]; tensor x_189_cast_fp16 = mul(x = var_5510_cast_fp16, y = var_5517_cast_fp16)[name = string("x_189_cast_fp16")]; tensor input_149_gamma_0_to_fp16 = const()[name = string("input_149_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(460763456)))]; tensor input_149_beta_0_to_fp16 = const()[name = string("input_149_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(460765568)))]; fp16 input_149_epsilon_0_to_fp16 = const()[name = string("input_149_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_149_cast_fp16 = batch_norm(beta = input_149_beta_0_to_fp16, epsilon = input_149_epsilon_0_to_fp16, gamma = input_149_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_189_cast_fp16)[name = string("input_149_cast_fp16")]; string x_191_pad_type_0 = const()[name = string("x_191_pad_type_0"), val = string("valid")]; tensor x_191_strides_0 = const()[name = string("x_191_strides_0"), val = tensor([1, 1])]; tensor x_191_pad_0 = const()[name = string("x_191_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_191_dilations_0 = const()[name = string("x_191_dilations_0"), val = tensor([1, 1])]; int32 x_191_groups_0 = const()[name = string("x_191_groups_0"), val = int32(1)]; tensor layers_17_fc1_weight_to_fp16 = const()[name = string("layers_17_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(460767680)))]; tensor layers_17_fc1_bias_to_fp16 = const()[name = string("layers_17_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(469156352)))]; tensor x_191_cast_fp16 = conv(bias = layers_17_fc1_bias_to_fp16, dilations = x_191_dilations_0, groups = x_191_groups_0, pad = x_191_pad_0, pad_type = x_191_pad_type_0, strides = x_191_strides_0, weight = layers_17_fc1_weight_to_fp16, x = input_149_cast_fp16)[name = string("x_191_cast_fp16")]; fp16 var_5532_to_fp16 = const()[name = string("op_5532_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_5533_cast_fp16 = mul(x = x_191_cast_fp16, y = var_5532_to_fp16)[name = string("op_5533_cast_fp16")]; tensor var_5534_cast_fp16 = mul(x = var_5533_cast_fp16, y = x_191_cast_fp16)[name = string("op_5534_cast_fp16")]; tensor var_5535_cast_fp16 = mul(x = var_5534_cast_fp16, y = x_191_cast_fp16)[name = string("op_5535_cast_fp16")]; tensor var_5536_cast_fp16 = add(x = x_191_cast_fp16, y = var_5535_cast_fp16)[name = string("op_5536_cast_fp16")]; fp16 var_5537_to_fp16 = const()[name = string("op_5537_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_41_cast_fp16 = mul(x = var_5536_cast_fp16, y = var_5537_to_fp16)[name = string("u_41_cast_fp16")]; fp16 var_5539_to_fp16 = const()[name = string("op_5539_to_fp16"), val = fp16(0x1p-1)]; tensor var_5540_cast_fp16 = mul(x = x_191_cast_fp16, y = var_5539_to_fp16)[name = string("op_5540_cast_fp16")]; tensor var_5541_cast_fp16 = tanh(x = u_41_cast_fp16)[name = string("op_5541_cast_fp16")]; fp16 var_5542_to_fp16 = const()[name = string("op_5542_to_fp16"), val = fp16(0x1p+0)]; tensor var_5543_cast_fp16 = add(x = var_5541_cast_fp16, y = var_5542_to_fp16)[name = string("op_5543_cast_fp16")]; tensor input_151_cast_fp16 = mul(x = var_5540_cast_fp16, y = var_5543_cast_fp16)[name = string("input_151_cast_fp16")]; string h_35_pad_type_0 = const()[name = string("h_35_pad_type_0"), val = string("valid")]; tensor h_35_strides_0 = const()[name = string("h_35_strides_0"), val = tensor([1, 1])]; tensor h_35_pad_0 = const()[name = string("h_35_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_35_dilations_0 = const()[name = string("h_35_dilations_0"), val = tensor([1, 1])]; int32 h_35_groups_0 = const()[name = string("h_35_groups_0"), val = int32(1)]; tensor layers_17_fc2_weight_to_fp16 = const()[name = string("layers_17_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(469164608)))]; tensor layers_17_fc2_bias_to_fp16 = const()[name = string("layers_17_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(477553280)))]; tensor h_35_cast_fp16 = conv(bias = layers_17_fc2_bias_to_fp16, dilations = h_35_dilations_0, groups = h_35_groups_0, pad = h_35_pad_0, pad_type = h_35_pad_type_0, strides = h_35_strides_0, weight = layers_17_fc2_weight_to_fp16, x = input_151_cast_fp16)[name = string("h_35_cast_fp16")]; tensor x_193_cast_fp16 = add(x = x_187_cast_fp16, y = h_35_cast_fp16)[name = string("x_193_cast_fp16")]; int32 var_5559 = const()[name = string("op_5559"), val = int32(1)]; tensor mu_73_axes_0 = const()[name = string("mu_73_axes_0"), val = tensor([1])]; bool mu_73_keep_dims_0 = const()[name = string("mu_73_keep_dims_0"), val = bool(true)]; tensor mu_73_cast_fp16 = reduce_mean(axes = mu_73_axes_0, keep_dims = mu_73_keep_dims_0, x = x_193_cast_fp16)[name = string("mu_73_cast_fp16")]; tensor var_5573_cast_fp16 = sub(x = x_193_cast_fp16, y = mu_73_cast_fp16)[name = string("op_5573_cast_fp16")]; fp16 var_5562_promoted_to_fp16 = const()[name = string("op_5562_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_5574_cast_fp16 = pow(x = var_5573_cast_fp16, y = var_5562_promoted_to_fp16)[name = string("op_5574_cast_fp16")]; tensor var_73_axes_0 = const()[name = string("var_73_axes_0"), val = tensor([1])]; bool var_73_keep_dims_0 = const()[name = string("var_73_keep_dims_0"), val = bool(true)]; tensor var_73_cast_fp16 = reduce_mean(axes = var_73_axes_0, keep_dims = var_73_keep_dims_0, x = var_5574_cast_fp16)[name = string("var_73_cast_fp16")]; fp16 var_5578_to_fp16 = const()[name = string("op_5578_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_5579_cast_fp16 = add(x = var_73_cast_fp16, y = var_5578_to_fp16)[name = string("op_5579_cast_fp16")]; fp32 var_5580_epsilon_0 = const()[name = string("op_5580_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_5580_cast_fp16 = rsqrt(epsilon = var_5580_epsilon_0, x = var_5579_cast_fp16)[name = string("op_5580_cast_fp16")]; tensor x_195_cast_fp16 = mul(x = var_5573_cast_fp16, y = var_5580_cast_fp16)[name = string("x_195_cast_fp16")]; tensor input_153_gamma_0_to_fp16 = const()[name = string("input_153_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(477555392)))]; tensor input_153_beta_0_to_fp16 = const()[name = string("input_153_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(477557504)))]; fp16 input_153_epsilon_0_to_fp16 = const()[name = string("input_153_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_153_cast_fp16 = batch_norm(beta = input_153_beta_0_to_fp16, epsilon = input_153_epsilon_0_to_fp16, gamma = input_153_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_195_cast_fp16)[name = string("input_153_cast_fp16")]; string var_5598_pad_type_0 = const()[name = string("op_5598_pad_type_0"), val = string("valid")]; tensor var_5598_strides_0 = const()[name = string("op_5598_strides_0"), val = tensor([1, 1])]; tensor var_5598_pad_0 = const()[name = string("op_5598_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5598_dilations_0 = const()[name = string("op_5598_dilations_0"), val = tensor([1, 1])]; int32 var_5598_groups_0 = const()[name = string("op_5598_groups_0"), val = int32(1)]; tensor var_5600_weight_0_to_fp16 = const()[name = string("op_5600_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(477559616)))]; tensor var_5600_bias_0_to_fp16 = const()[name = string("op_5600_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(479656832)))]; tensor var_5600_cast_fp16 = conv(bias = var_5600_bias_0_to_fp16, dilations = var_5598_dilations_0, groups = var_5598_groups_0, pad = var_5598_pad_0, pad_type = var_5598_pad_type_0, strides = var_5598_strides_0, weight = var_5600_weight_0_to_fp16, x = input_153_cast_fp16)[name = string("op_5600_cast_fp16")]; string var_5607_pad_type_0 = const()[name = string("op_5607_pad_type_0"), val = string("valid")]; tensor var_5607_strides_0 = const()[name = string("op_5607_strides_0"), val = tensor([1, 1])]; tensor var_5607_pad_0 = const()[name = string("op_5607_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5607_dilations_0 = const()[name = string("op_5607_dilations_0"), val = tensor([1, 1])]; int32 var_5607_groups_0 = const()[name = string("op_5607_groups_0"), val = int32(1)]; tensor layers_18_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_18_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(479658944)))]; tensor layers_18_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_18_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(481756160)))]; tensor var_5607_cast_fp16 = conv(bias = layers_18_self_attn_k_proj_bias_to_fp16, dilations = var_5607_dilations_0, groups = var_5607_groups_0, pad = var_5607_pad_0, pad_type = var_5607_pad_type_0, strides = var_5607_strides_0, weight = layers_18_self_attn_k_proj_weight_to_fp16, x = input_153_cast_fp16)[name = string("op_5607_cast_fp16")]; string var_5614_pad_type_0 = const()[name = string("op_5614_pad_type_0"), val = string("valid")]; tensor var_5614_strides_0 = const()[name = string("op_5614_strides_0"), val = tensor([1, 1])]; tensor var_5614_pad_0 = const()[name = string("op_5614_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5614_dilations_0 = const()[name = string("op_5614_dilations_0"), val = tensor([1, 1])]; int32 var_5614_groups_0 = const()[name = string("op_5614_groups_0"), val = int32(1)]; tensor layers_18_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_18_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(481758272)))]; tensor layers_18_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_18_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(483855488)))]; tensor var_5614_cast_fp16 = conv(bias = layers_18_self_attn_v_proj_bias_to_fp16, dilations = var_5614_dilations_0, groups = var_5614_groups_0, pad = var_5614_pad_0, pad_type = var_5614_pad_type_0, strides = var_5614_strides_0, weight = layers_18_self_attn_v_proj_weight_to_fp16, x = input_153_cast_fp16)[name = string("op_5614_cast_fp16")]; tensor tile_54 = const()[name = string("tile_54"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(483857600)))]; int32 var_5615_axis_0 = const()[name = string("op_5615_axis_0"), val = int32(1)]; tensor var_5615_cast_fp16_0, tensor var_5615_cast_fp16_1, tensor var_5615_cast_fp16_2, tensor var_5615_cast_fp16_3, tensor var_5615_cast_fp16_4, tensor var_5615_cast_fp16_5, tensor var_5615_cast_fp16_6, tensor var_5615_cast_fp16_7, tensor var_5615_cast_fp16_8, tensor var_5615_cast_fp16_9, tensor var_5615_cast_fp16_10, tensor var_5615_cast_fp16_11, tensor var_5615_cast_fp16_12, tensor var_5615_cast_fp16_13, tensor var_5615_cast_fp16_14, tensor var_5615_cast_fp16_15 = split(axis = var_5615_axis_0, split_sizes = tile_54, x = var_5600_cast_fp16)[name = string("op_5615_cast_fp16")]; tensor tile_55 = const()[name = string("tile_55"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(483857728)))]; int32 var_5632_axis_0 = const()[name = string("op_5632_axis_0"), val = int32(1)]; tensor var_5632_cast_fp16_0, tensor var_5632_cast_fp16_1, tensor var_5632_cast_fp16_2, tensor var_5632_cast_fp16_3, tensor var_5632_cast_fp16_4, tensor var_5632_cast_fp16_5, tensor var_5632_cast_fp16_6, tensor var_5632_cast_fp16_7, tensor var_5632_cast_fp16_8, tensor var_5632_cast_fp16_9, tensor var_5632_cast_fp16_10, tensor var_5632_cast_fp16_11, tensor var_5632_cast_fp16_12, tensor var_5632_cast_fp16_13, tensor var_5632_cast_fp16_14, tensor var_5632_cast_fp16_15 = split(axis = var_5632_axis_0, split_sizes = tile_55, x = var_5607_cast_fp16)[name = string("op_5632_cast_fp16")]; tensor tile_56 = const()[name = string("tile_56"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(483857856)))]; int32 var_5649_axis_0 = const()[name = string("op_5649_axis_0"), val = int32(1)]; tensor var_5649_cast_fp16_0, tensor var_5649_cast_fp16_1, tensor var_5649_cast_fp16_2, tensor var_5649_cast_fp16_3, tensor var_5649_cast_fp16_4, tensor var_5649_cast_fp16_5, tensor var_5649_cast_fp16_6, tensor var_5649_cast_fp16_7, tensor var_5649_cast_fp16_8, tensor var_5649_cast_fp16_9, tensor var_5649_cast_fp16_10, tensor var_5649_cast_fp16_11, tensor var_5649_cast_fp16_12, tensor var_5649_cast_fp16_13, tensor var_5649_cast_fp16_14, tensor var_5649_cast_fp16_15 = split(axis = var_5649_axis_0, split_sizes = tile_56, x = var_5614_cast_fp16)[name = string("op_5649_cast_fp16")]; tensor transpose_576_perm_0 = const()[name = string("transpose_576_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2884 = const()[name = string("concat_2884"), val = tensor([1, 104, 64])]; tensor transpose_576_cast_fp16 = transpose(perm = transpose_576_perm_0, x = var_5615_cast_fp16_0)[name = string("transpose_3359")]; tensor reshape_864_cast_fp16 = reshape(shape = concat_2884, x = transpose_576_cast_fp16)[name = string("reshape_864_cast_fp16")]; tensor transpose_577_perm_0 = const()[name = string("transpose_577_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2885 = const()[name = string("concat_2885"), val = tensor([1, 64, 104])]; tensor transpose_577_cast_fp16 = transpose(perm = transpose_577_perm_0, x = var_5632_cast_fp16_0)[name = string("transpose_3358")]; tensor reshape_865_cast_fp16 = reshape(shape = concat_2885, x = transpose_577_cast_fp16)[name = string("reshape_865_cast_fp16")]; bool matmul_288_transpose_x_0 = const()[name = string("matmul_288_transpose_x_0"), val = bool(false)]; bool matmul_288_transpose_y_0 = const()[name = string("matmul_288_transpose_y_0"), val = bool(false)]; tensor matmul_288_cast_fp16 = matmul(transpose_x = matmul_288_transpose_x_0, transpose_y = matmul_288_transpose_y_0, x = reshape_864_cast_fp16, y = reshape_865_cast_fp16)[name = string("matmul_288_cast_fp16")]; tensor concat_2889 = const()[name = string("concat_2889"), val = tensor([1, 1, 104, 104])]; tensor reshape_866_cast_fp16 = reshape(shape = concat_2889, x = matmul_288_cast_fp16)[name = string("reshape_866_cast_fp16")]; tensor transpose_2976_perm_0 = const()[name = string("transpose_2976_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2976 = transpose(perm = transpose_2976_perm_0, x = reshape_866_cast_fp16)[name = string("transpose_3357")]; tensor w_1155_cast_fp16 = add(x = transpose_2976, y = transpose_2305)[name = string("w_1155_cast_fp16")]; tensor var_5671_cast_fp16 = softmax(axis = var_5559, x = w_1155_cast_fp16)[name = string("op_5671_cast_fp16")]; string var_5673_equation_0 = const()[name = string("op_5673_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5673_cast_fp16 = einsum(equation = var_5673_equation_0, values = (var_5649_cast_fp16_0, var_5671_cast_fp16))[name = string("op_5673_cast_fp16")]; tensor transpose_578_perm_0 = const()[name = string("transpose_578_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2894 = const()[name = string("concat_2894"), val = tensor([1, 104, 64])]; tensor transpose_578_cast_fp16 = transpose(perm = transpose_578_perm_0, x = var_5615_cast_fp16_1)[name = string("transpose_3356")]; tensor reshape_867_cast_fp16 = reshape(shape = concat_2894, x = transpose_578_cast_fp16)[name = string("reshape_867_cast_fp16")]; tensor transpose_579_perm_0 = const()[name = string("transpose_579_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2895 = const()[name = string("concat_2895"), val = tensor([1, 64, 104])]; tensor transpose_579_cast_fp16 = transpose(perm = transpose_579_perm_0, x = var_5632_cast_fp16_1)[name = string("transpose_3355")]; tensor reshape_868_cast_fp16 = reshape(shape = concat_2895, x = transpose_579_cast_fp16)[name = string("reshape_868_cast_fp16")]; bool matmul_289_transpose_x_0 = const()[name = string("matmul_289_transpose_x_0"), val = bool(false)]; bool matmul_289_transpose_y_0 = const()[name = string("matmul_289_transpose_y_0"), val = bool(false)]; tensor matmul_289_cast_fp16 = matmul(transpose_x = matmul_289_transpose_x_0, transpose_y = matmul_289_transpose_y_0, x = reshape_867_cast_fp16, y = reshape_868_cast_fp16)[name = string("matmul_289_cast_fp16")]; tensor concat_2899 = const()[name = string("concat_2899"), val = tensor([1, 1, 104, 104])]; tensor reshape_869_cast_fp16 = reshape(shape = concat_2899, x = matmul_289_cast_fp16)[name = string("reshape_869_cast_fp16")]; tensor transpose_2977_perm_0 = const()[name = string("transpose_2977_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2977 = transpose(perm = transpose_2977_perm_0, x = reshape_869_cast_fp16)[name = string("transpose_3354")]; tensor w_1159_cast_fp16 = add(x = transpose_2977, y = transpose_2305)[name = string("w_1159_cast_fp16")]; tensor var_5679_cast_fp16 = softmax(axis = var_5559, x = w_1159_cast_fp16)[name = string("op_5679_cast_fp16")]; string var_5681_equation_0 = const()[name = string("op_5681_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5681_cast_fp16 = einsum(equation = var_5681_equation_0, values = (var_5649_cast_fp16_1, var_5679_cast_fp16))[name = string("op_5681_cast_fp16")]; tensor transpose_580_perm_0 = const()[name = string("transpose_580_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2904 = const()[name = string("concat_2904"), val = tensor([1, 104, 64])]; tensor transpose_580_cast_fp16 = transpose(perm = transpose_580_perm_0, x = var_5615_cast_fp16_2)[name = string("transpose_3353")]; tensor reshape_870_cast_fp16 = reshape(shape = concat_2904, x = transpose_580_cast_fp16)[name = string("reshape_870_cast_fp16")]; tensor transpose_581_perm_0 = const()[name = string("transpose_581_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2905 = const()[name = string("concat_2905"), val = tensor([1, 64, 104])]; tensor transpose_581_cast_fp16 = transpose(perm = transpose_581_perm_0, x = var_5632_cast_fp16_2)[name = string("transpose_3352")]; tensor reshape_871_cast_fp16 = reshape(shape = concat_2905, x = transpose_581_cast_fp16)[name = string("reshape_871_cast_fp16")]; bool matmul_290_transpose_x_0 = const()[name = string("matmul_290_transpose_x_0"), val = bool(false)]; bool matmul_290_transpose_y_0 = const()[name = string("matmul_290_transpose_y_0"), val = bool(false)]; tensor matmul_290_cast_fp16 = matmul(transpose_x = matmul_290_transpose_x_0, transpose_y = matmul_290_transpose_y_0, x = reshape_870_cast_fp16, y = reshape_871_cast_fp16)[name = string("matmul_290_cast_fp16")]; tensor concat_2909 = const()[name = string("concat_2909"), val = tensor([1, 1, 104, 104])]; tensor reshape_872_cast_fp16 = reshape(shape = concat_2909, x = matmul_290_cast_fp16)[name = string("reshape_872_cast_fp16")]; tensor transpose_2978_perm_0 = const()[name = string("transpose_2978_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2978 = transpose(perm = transpose_2978_perm_0, x = reshape_872_cast_fp16)[name = string("transpose_3351")]; tensor w_1163_cast_fp16 = add(x = transpose_2978, y = transpose_2305)[name = string("w_1163_cast_fp16")]; tensor var_5687_cast_fp16 = softmax(axis = var_5559, x = w_1163_cast_fp16)[name = string("op_5687_cast_fp16")]; string var_5689_equation_0 = const()[name = string("op_5689_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5689_cast_fp16 = einsum(equation = var_5689_equation_0, values = (var_5649_cast_fp16_2, var_5687_cast_fp16))[name = string("op_5689_cast_fp16")]; tensor transpose_582_perm_0 = const()[name = string("transpose_582_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2914 = const()[name = string("concat_2914"), val = tensor([1, 104, 64])]; tensor transpose_582_cast_fp16 = transpose(perm = transpose_582_perm_0, x = var_5615_cast_fp16_3)[name = string("transpose_3350")]; tensor reshape_873_cast_fp16 = reshape(shape = concat_2914, x = transpose_582_cast_fp16)[name = string("reshape_873_cast_fp16")]; tensor transpose_583_perm_0 = const()[name = string("transpose_583_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2915 = const()[name = string("concat_2915"), val = tensor([1, 64, 104])]; tensor transpose_583_cast_fp16 = transpose(perm = transpose_583_perm_0, x = var_5632_cast_fp16_3)[name = string("transpose_3349")]; tensor reshape_874_cast_fp16 = reshape(shape = concat_2915, x = transpose_583_cast_fp16)[name = string("reshape_874_cast_fp16")]; bool matmul_291_transpose_x_0 = const()[name = string("matmul_291_transpose_x_0"), val = bool(false)]; bool matmul_291_transpose_y_0 = const()[name = string("matmul_291_transpose_y_0"), val = bool(false)]; tensor matmul_291_cast_fp16 = matmul(transpose_x = matmul_291_transpose_x_0, transpose_y = matmul_291_transpose_y_0, x = reshape_873_cast_fp16, y = reshape_874_cast_fp16)[name = string("matmul_291_cast_fp16")]; tensor concat_2919 = const()[name = string("concat_2919"), val = tensor([1, 1, 104, 104])]; tensor reshape_875_cast_fp16 = reshape(shape = concat_2919, x = matmul_291_cast_fp16)[name = string("reshape_875_cast_fp16")]; tensor transpose_2979_perm_0 = const()[name = string("transpose_2979_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2979 = transpose(perm = transpose_2979_perm_0, x = reshape_875_cast_fp16)[name = string("transpose_3348")]; tensor w_1167_cast_fp16 = add(x = transpose_2979, y = transpose_2305)[name = string("w_1167_cast_fp16")]; tensor var_5695_cast_fp16 = softmax(axis = var_5559, x = w_1167_cast_fp16)[name = string("op_5695_cast_fp16")]; string var_5697_equation_0 = const()[name = string("op_5697_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5697_cast_fp16 = einsum(equation = var_5697_equation_0, values = (var_5649_cast_fp16_3, var_5695_cast_fp16))[name = string("op_5697_cast_fp16")]; tensor transpose_584_perm_0 = const()[name = string("transpose_584_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2924 = const()[name = string("concat_2924"), val = tensor([1, 104, 64])]; tensor transpose_584_cast_fp16 = transpose(perm = transpose_584_perm_0, x = var_5615_cast_fp16_4)[name = string("transpose_3347")]; tensor reshape_876_cast_fp16 = reshape(shape = concat_2924, x = transpose_584_cast_fp16)[name = string("reshape_876_cast_fp16")]; tensor transpose_585_perm_0 = const()[name = string("transpose_585_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2925 = const()[name = string("concat_2925"), val = tensor([1, 64, 104])]; tensor transpose_585_cast_fp16 = transpose(perm = transpose_585_perm_0, x = var_5632_cast_fp16_4)[name = string("transpose_3346")]; tensor reshape_877_cast_fp16 = reshape(shape = concat_2925, x = transpose_585_cast_fp16)[name = string("reshape_877_cast_fp16")]; bool matmul_292_transpose_x_0 = const()[name = string("matmul_292_transpose_x_0"), val = bool(false)]; bool matmul_292_transpose_y_0 = const()[name = string("matmul_292_transpose_y_0"), val = bool(false)]; tensor matmul_292_cast_fp16 = matmul(transpose_x = matmul_292_transpose_x_0, transpose_y = matmul_292_transpose_y_0, x = reshape_876_cast_fp16, y = reshape_877_cast_fp16)[name = string("matmul_292_cast_fp16")]; tensor concat_2929 = const()[name = string("concat_2929"), val = tensor([1, 1, 104, 104])]; tensor reshape_878_cast_fp16 = reshape(shape = concat_2929, x = matmul_292_cast_fp16)[name = string("reshape_878_cast_fp16")]; tensor transpose_2980_perm_0 = const()[name = string("transpose_2980_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2980 = transpose(perm = transpose_2980_perm_0, x = reshape_878_cast_fp16)[name = string("transpose_3345")]; tensor w_1171_cast_fp16 = add(x = transpose_2980, y = transpose_2305)[name = string("w_1171_cast_fp16")]; tensor var_5703_cast_fp16 = softmax(axis = var_5559, x = w_1171_cast_fp16)[name = string("op_5703_cast_fp16")]; string var_5705_equation_0 = const()[name = string("op_5705_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5705_cast_fp16 = einsum(equation = var_5705_equation_0, values = (var_5649_cast_fp16_4, var_5703_cast_fp16))[name = string("op_5705_cast_fp16")]; tensor transpose_586_perm_0 = const()[name = string("transpose_586_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2934 = const()[name = string("concat_2934"), val = tensor([1, 104, 64])]; tensor transpose_586_cast_fp16 = transpose(perm = transpose_586_perm_0, x = var_5615_cast_fp16_5)[name = string("transpose_3344")]; tensor reshape_879_cast_fp16 = reshape(shape = concat_2934, x = transpose_586_cast_fp16)[name = string("reshape_879_cast_fp16")]; tensor transpose_587_perm_0 = const()[name = string("transpose_587_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2935 = const()[name = string("concat_2935"), val = tensor([1, 64, 104])]; tensor transpose_587_cast_fp16 = transpose(perm = transpose_587_perm_0, x = var_5632_cast_fp16_5)[name = string("transpose_3343")]; tensor reshape_880_cast_fp16 = reshape(shape = concat_2935, x = transpose_587_cast_fp16)[name = string("reshape_880_cast_fp16")]; bool matmul_293_transpose_x_0 = const()[name = string("matmul_293_transpose_x_0"), val = bool(false)]; bool matmul_293_transpose_y_0 = const()[name = string("matmul_293_transpose_y_0"), val = bool(false)]; tensor matmul_293_cast_fp16 = matmul(transpose_x = matmul_293_transpose_x_0, transpose_y = matmul_293_transpose_y_0, x = reshape_879_cast_fp16, y = reshape_880_cast_fp16)[name = string("matmul_293_cast_fp16")]; tensor concat_2939 = const()[name = string("concat_2939"), val = tensor([1, 1, 104, 104])]; tensor reshape_881_cast_fp16 = reshape(shape = concat_2939, x = matmul_293_cast_fp16)[name = string("reshape_881_cast_fp16")]; tensor transpose_2981_perm_0 = const()[name = string("transpose_2981_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2981 = transpose(perm = transpose_2981_perm_0, x = reshape_881_cast_fp16)[name = string("transpose_3342")]; tensor w_1175_cast_fp16 = add(x = transpose_2981, y = transpose_2305)[name = string("w_1175_cast_fp16")]; tensor var_5711_cast_fp16 = softmax(axis = var_5559, x = w_1175_cast_fp16)[name = string("op_5711_cast_fp16")]; string var_5713_equation_0 = const()[name = string("op_5713_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5713_cast_fp16 = einsum(equation = var_5713_equation_0, values = (var_5649_cast_fp16_5, var_5711_cast_fp16))[name = string("op_5713_cast_fp16")]; tensor transpose_588_perm_0 = const()[name = string("transpose_588_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2944 = const()[name = string("concat_2944"), val = tensor([1, 104, 64])]; tensor transpose_588_cast_fp16 = transpose(perm = transpose_588_perm_0, x = var_5615_cast_fp16_6)[name = string("transpose_3341")]; tensor reshape_882_cast_fp16 = reshape(shape = concat_2944, x = transpose_588_cast_fp16)[name = string("reshape_882_cast_fp16")]; tensor transpose_589_perm_0 = const()[name = string("transpose_589_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2945 = const()[name = string("concat_2945"), val = tensor([1, 64, 104])]; tensor transpose_589_cast_fp16 = transpose(perm = transpose_589_perm_0, x = var_5632_cast_fp16_6)[name = string("transpose_3340")]; tensor reshape_883_cast_fp16 = reshape(shape = concat_2945, x = transpose_589_cast_fp16)[name = string("reshape_883_cast_fp16")]; bool matmul_294_transpose_x_0 = const()[name = string("matmul_294_transpose_x_0"), val = bool(false)]; bool matmul_294_transpose_y_0 = const()[name = string("matmul_294_transpose_y_0"), val = bool(false)]; tensor matmul_294_cast_fp16 = matmul(transpose_x = matmul_294_transpose_x_0, transpose_y = matmul_294_transpose_y_0, x = reshape_882_cast_fp16, y = reshape_883_cast_fp16)[name = string("matmul_294_cast_fp16")]; tensor concat_2949 = const()[name = string("concat_2949"), val = tensor([1, 1, 104, 104])]; tensor reshape_884_cast_fp16 = reshape(shape = concat_2949, x = matmul_294_cast_fp16)[name = string("reshape_884_cast_fp16")]; tensor transpose_2982_perm_0 = const()[name = string("transpose_2982_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2982 = transpose(perm = transpose_2982_perm_0, x = reshape_884_cast_fp16)[name = string("transpose_3339")]; tensor w_1179_cast_fp16 = add(x = transpose_2982, y = transpose_2305)[name = string("w_1179_cast_fp16")]; tensor var_5719_cast_fp16 = softmax(axis = var_5559, x = w_1179_cast_fp16)[name = string("op_5719_cast_fp16")]; string var_5721_equation_0 = const()[name = string("op_5721_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5721_cast_fp16 = einsum(equation = var_5721_equation_0, values = (var_5649_cast_fp16_6, var_5719_cast_fp16))[name = string("op_5721_cast_fp16")]; tensor transpose_590_perm_0 = const()[name = string("transpose_590_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2954 = const()[name = string("concat_2954"), val = tensor([1, 104, 64])]; tensor transpose_590_cast_fp16 = transpose(perm = transpose_590_perm_0, x = var_5615_cast_fp16_7)[name = string("transpose_3338")]; tensor reshape_885_cast_fp16 = reshape(shape = concat_2954, x = transpose_590_cast_fp16)[name = string("reshape_885_cast_fp16")]; tensor transpose_591_perm_0 = const()[name = string("transpose_591_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2955 = const()[name = string("concat_2955"), val = tensor([1, 64, 104])]; tensor transpose_591_cast_fp16 = transpose(perm = transpose_591_perm_0, x = var_5632_cast_fp16_7)[name = string("transpose_3337")]; tensor reshape_886_cast_fp16 = reshape(shape = concat_2955, x = transpose_591_cast_fp16)[name = string("reshape_886_cast_fp16")]; bool matmul_295_transpose_x_0 = const()[name = string("matmul_295_transpose_x_0"), val = bool(false)]; bool matmul_295_transpose_y_0 = const()[name = string("matmul_295_transpose_y_0"), val = bool(false)]; tensor matmul_295_cast_fp16 = matmul(transpose_x = matmul_295_transpose_x_0, transpose_y = matmul_295_transpose_y_0, x = reshape_885_cast_fp16, y = reshape_886_cast_fp16)[name = string("matmul_295_cast_fp16")]; tensor concat_2959 = const()[name = string("concat_2959"), val = tensor([1, 1, 104, 104])]; tensor reshape_887_cast_fp16 = reshape(shape = concat_2959, x = matmul_295_cast_fp16)[name = string("reshape_887_cast_fp16")]; tensor transpose_2983_perm_0 = const()[name = string("transpose_2983_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2983 = transpose(perm = transpose_2983_perm_0, x = reshape_887_cast_fp16)[name = string("transpose_3336")]; tensor w_1183_cast_fp16 = add(x = transpose_2983, y = transpose_2305)[name = string("w_1183_cast_fp16")]; tensor var_5727_cast_fp16 = softmax(axis = var_5559, x = w_1183_cast_fp16)[name = string("op_5727_cast_fp16")]; string var_5729_equation_0 = const()[name = string("op_5729_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5729_cast_fp16 = einsum(equation = var_5729_equation_0, values = (var_5649_cast_fp16_7, var_5727_cast_fp16))[name = string("op_5729_cast_fp16")]; tensor transpose_592_perm_0 = const()[name = string("transpose_592_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2964 = const()[name = string("concat_2964"), val = tensor([1, 104, 64])]; tensor transpose_592_cast_fp16 = transpose(perm = transpose_592_perm_0, x = var_5615_cast_fp16_8)[name = string("transpose_3335")]; tensor reshape_888_cast_fp16 = reshape(shape = concat_2964, x = transpose_592_cast_fp16)[name = string("reshape_888_cast_fp16")]; tensor transpose_593_perm_0 = const()[name = string("transpose_593_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2965 = const()[name = string("concat_2965"), val = tensor([1, 64, 104])]; tensor transpose_593_cast_fp16 = transpose(perm = transpose_593_perm_0, x = var_5632_cast_fp16_8)[name = string("transpose_3334")]; tensor reshape_889_cast_fp16 = reshape(shape = concat_2965, x = transpose_593_cast_fp16)[name = string("reshape_889_cast_fp16")]; bool matmul_296_transpose_x_0 = const()[name = string("matmul_296_transpose_x_0"), val = bool(false)]; bool matmul_296_transpose_y_0 = const()[name = string("matmul_296_transpose_y_0"), val = bool(false)]; tensor matmul_296_cast_fp16 = matmul(transpose_x = matmul_296_transpose_x_0, transpose_y = matmul_296_transpose_y_0, x = reshape_888_cast_fp16, y = reshape_889_cast_fp16)[name = string("matmul_296_cast_fp16")]; tensor concat_2969 = const()[name = string("concat_2969"), val = tensor([1, 1, 104, 104])]; tensor reshape_890_cast_fp16 = reshape(shape = concat_2969, x = matmul_296_cast_fp16)[name = string("reshape_890_cast_fp16")]; tensor transpose_2984_perm_0 = const()[name = string("transpose_2984_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2984 = transpose(perm = transpose_2984_perm_0, x = reshape_890_cast_fp16)[name = string("transpose_3333")]; tensor w_1187_cast_fp16 = add(x = transpose_2984, y = transpose_2305)[name = string("w_1187_cast_fp16")]; tensor var_5735_cast_fp16 = softmax(axis = var_5559, x = w_1187_cast_fp16)[name = string("op_5735_cast_fp16")]; string var_5737_equation_0 = const()[name = string("op_5737_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5737_cast_fp16 = einsum(equation = var_5737_equation_0, values = (var_5649_cast_fp16_8, var_5735_cast_fp16))[name = string("op_5737_cast_fp16")]; tensor transpose_594_perm_0 = const()[name = string("transpose_594_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2974 = const()[name = string("concat_2974"), val = tensor([1, 104, 64])]; tensor transpose_594_cast_fp16 = transpose(perm = transpose_594_perm_0, x = var_5615_cast_fp16_9)[name = string("transpose_3332")]; tensor reshape_891_cast_fp16 = reshape(shape = concat_2974, x = transpose_594_cast_fp16)[name = string("reshape_891_cast_fp16")]; tensor transpose_595_perm_0 = const()[name = string("transpose_595_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2975 = const()[name = string("concat_2975"), val = tensor([1, 64, 104])]; tensor transpose_595_cast_fp16 = transpose(perm = transpose_595_perm_0, x = var_5632_cast_fp16_9)[name = string("transpose_3331")]; tensor reshape_892_cast_fp16 = reshape(shape = concat_2975, x = transpose_595_cast_fp16)[name = string("reshape_892_cast_fp16")]; bool matmul_297_transpose_x_0 = const()[name = string("matmul_297_transpose_x_0"), val = bool(false)]; bool matmul_297_transpose_y_0 = const()[name = string("matmul_297_transpose_y_0"), val = bool(false)]; tensor matmul_297_cast_fp16 = matmul(transpose_x = matmul_297_transpose_x_0, transpose_y = matmul_297_transpose_y_0, x = reshape_891_cast_fp16, y = reshape_892_cast_fp16)[name = string("matmul_297_cast_fp16")]; tensor concat_2979 = const()[name = string("concat_2979"), val = tensor([1, 1, 104, 104])]; tensor reshape_893_cast_fp16 = reshape(shape = concat_2979, x = matmul_297_cast_fp16)[name = string("reshape_893_cast_fp16")]; tensor transpose_2985_perm_0 = const()[name = string("transpose_2985_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2985 = transpose(perm = transpose_2985_perm_0, x = reshape_893_cast_fp16)[name = string("transpose_3330")]; tensor w_1191_cast_fp16 = add(x = transpose_2985, y = transpose_2305)[name = string("w_1191_cast_fp16")]; tensor var_5743_cast_fp16 = softmax(axis = var_5559, x = w_1191_cast_fp16)[name = string("op_5743_cast_fp16")]; string var_5745_equation_0 = const()[name = string("op_5745_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5745_cast_fp16 = einsum(equation = var_5745_equation_0, values = (var_5649_cast_fp16_9, var_5743_cast_fp16))[name = string("op_5745_cast_fp16")]; tensor transpose_596_perm_0 = const()[name = string("transpose_596_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2984 = const()[name = string("concat_2984"), val = tensor([1, 104, 64])]; tensor transpose_596_cast_fp16 = transpose(perm = transpose_596_perm_0, x = var_5615_cast_fp16_10)[name = string("transpose_3329")]; tensor reshape_894_cast_fp16 = reshape(shape = concat_2984, x = transpose_596_cast_fp16)[name = string("reshape_894_cast_fp16")]; tensor transpose_597_perm_0 = const()[name = string("transpose_597_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2985 = const()[name = string("concat_2985"), val = tensor([1, 64, 104])]; tensor transpose_597_cast_fp16 = transpose(perm = transpose_597_perm_0, x = var_5632_cast_fp16_10)[name = string("transpose_3328")]; tensor reshape_895_cast_fp16 = reshape(shape = concat_2985, x = transpose_597_cast_fp16)[name = string("reshape_895_cast_fp16")]; bool matmul_298_transpose_x_0 = const()[name = string("matmul_298_transpose_x_0"), val = bool(false)]; bool matmul_298_transpose_y_0 = const()[name = string("matmul_298_transpose_y_0"), val = bool(false)]; tensor matmul_298_cast_fp16 = matmul(transpose_x = matmul_298_transpose_x_0, transpose_y = matmul_298_transpose_y_0, x = reshape_894_cast_fp16, y = reshape_895_cast_fp16)[name = string("matmul_298_cast_fp16")]; tensor concat_2989 = const()[name = string("concat_2989"), val = tensor([1, 1, 104, 104])]; tensor reshape_896_cast_fp16 = reshape(shape = concat_2989, x = matmul_298_cast_fp16)[name = string("reshape_896_cast_fp16")]; tensor transpose_2986_perm_0 = const()[name = string("transpose_2986_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2986 = transpose(perm = transpose_2986_perm_0, x = reshape_896_cast_fp16)[name = string("transpose_3327")]; tensor w_1195_cast_fp16 = add(x = transpose_2986, y = transpose_2305)[name = string("w_1195_cast_fp16")]; tensor var_5751_cast_fp16 = softmax(axis = var_5559, x = w_1195_cast_fp16)[name = string("op_5751_cast_fp16")]; string var_5753_equation_0 = const()[name = string("op_5753_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5753_cast_fp16 = einsum(equation = var_5753_equation_0, values = (var_5649_cast_fp16_10, var_5751_cast_fp16))[name = string("op_5753_cast_fp16")]; tensor transpose_598_perm_0 = const()[name = string("transpose_598_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_2994 = const()[name = string("concat_2994"), val = tensor([1, 104, 64])]; tensor transpose_598_cast_fp16 = transpose(perm = transpose_598_perm_0, x = var_5615_cast_fp16_11)[name = string("transpose_3326")]; tensor reshape_897_cast_fp16 = reshape(shape = concat_2994, x = transpose_598_cast_fp16)[name = string("reshape_897_cast_fp16")]; tensor transpose_599_perm_0 = const()[name = string("transpose_599_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_2995 = const()[name = string("concat_2995"), val = tensor([1, 64, 104])]; tensor transpose_599_cast_fp16 = transpose(perm = transpose_599_perm_0, x = var_5632_cast_fp16_11)[name = string("transpose_3325")]; tensor reshape_898_cast_fp16 = reshape(shape = concat_2995, x = transpose_599_cast_fp16)[name = string("reshape_898_cast_fp16")]; bool matmul_299_transpose_x_0 = const()[name = string("matmul_299_transpose_x_0"), val = bool(false)]; bool matmul_299_transpose_y_0 = const()[name = string("matmul_299_transpose_y_0"), val = bool(false)]; tensor matmul_299_cast_fp16 = matmul(transpose_x = matmul_299_transpose_x_0, transpose_y = matmul_299_transpose_y_0, x = reshape_897_cast_fp16, y = reshape_898_cast_fp16)[name = string("matmul_299_cast_fp16")]; tensor concat_2999 = const()[name = string("concat_2999"), val = tensor([1, 1, 104, 104])]; tensor reshape_899_cast_fp16 = reshape(shape = concat_2999, x = matmul_299_cast_fp16)[name = string("reshape_899_cast_fp16")]; tensor transpose_2987_perm_0 = const()[name = string("transpose_2987_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2987 = transpose(perm = transpose_2987_perm_0, x = reshape_899_cast_fp16)[name = string("transpose_3324")]; tensor w_1199_cast_fp16 = add(x = transpose_2987, y = transpose_2305)[name = string("w_1199_cast_fp16")]; tensor var_5759_cast_fp16 = softmax(axis = var_5559, x = w_1199_cast_fp16)[name = string("op_5759_cast_fp16")]; string var_5761_equation_0 = const()[name = string("op_5761_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5761_cast_fp16 = einsum(equation = var_5761_equation_0, values = (var_5649_cast_fp16_11, var_5759_cast_fp16))[name = string("op_5761_cast_fp16")]; tensor transpose_600_perm_0 = const()[name = string("transpose_600_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3004 = const()[name = string("concat_3004"), val = tensor([1, 104, 64])]; tensor transpose_600_cast_fp16 = transpose(perm = transpose_600_perm_0, x = var_5615_cast_fp16_12)[name = string("transpose_3323")]; tensor reshape_900_cast_fp16 = reshape(shape = concat_3004, x = transpose_600_cast_fp16)[name = string("reshape_900_cast_fp16")]; tensor transpose_601_perm_0 = const()[name = string("transpose_601_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3005 = const()[name = string("concat_3005"), val = tensor([1, 64, 104])]; tensor transpose_601_cast_fp16 = transpose(perm = transpose_601_perm_0, x = var_5632_cast_fp16_12)[name = string("transpose_3322")]; tensor reshape_901_cast_fp16 = reshape(shape = concat_3005, x = transpose_601_cast_fp16)[name = string("reshape_901_cast_fp16")]; bool matmul_300_transpose_x_0 = const()[name = string("matmul_300_transpose_x_0"), val = bool(false)]; bool matmul_300_transpose_y_0 = const()[name = string("matmul_300_transpose_y_0"), val = bool(false)]; tensor matmul_300_cast_fp16 = matmul(transpose_x = matmul_300_transpose_x_0, transpose_y = matmul_300_transpose_y_0, x = reshape_900_cast_fp16, y = reshape_901_cast_fp16)[name = string("matmul_300_cast_fp16")]; tensor concat_3009 = const()[name = string("concat_3009"), val = tensor([1, 1, 104, 104])]; tensor reshape_902_cast_fp16 = reshape(shape = concat_3009, x = matmul_300_cast_fp16)[name = string("reshape_902_cast_fp16")]; tensor transpose_2988_perm_0 = const()[name = string("transpose_2988_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2988 = transpose(perm = transpose_2988_perm_0, x = reshape_902_cast_fp16)[name = string("transpose_3321")]; tensor w_1203_cast_fp16 = add(x = transpose_2988, y = transpose_2305)[name = string("w_1203_cast_fp16")]; tensor var_5767_cast_fp16 = softmax(axis = var_5559, x = w_1203_cast_fp16)[name = string("op_5767_cast_fp16")]; string var_5769_equation_0 = const()[name = string("op_5769_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5769_cast_fp16 = einsum(equation = var_5769_equation_0, values = (var_5649_cast_fp16_12, var_5767_cast_fp16))[name = string("op_5769_cast_fp16")]; tensor transpose_602_perm_0 = const()[name = string("transpose_602_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3014 = const()[name = string("concat_3014"), val = tensor([1, 104, 64])]; tensor transpose_602_cast_fp16 = transpose(perm = transpose_602_perm_0, x = var_5615_cast_fp16_13)[name = string("transpose_3320")]; tensor reshape_903_cast_fp16 = reshape(shape = concat_3014, x = transpose_602_cast_fp16)[name = string("reshape_903_cast_fp16")]; tensor transpose_603_perm_0 = const()[name = string("transpose_603_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3015 = const()[name = string("concat_3015"), val = tensor([1, 64, 104])]; tensor transpose_603_cast_fp16 = transpose(perm = transpose_603_perm_0, x = var_5632_cast_fp16_13)[name = string("transpose_3319")]; tensor reshape_904_cast_fp16 = reshape(shape = concat_3015, x = transpose_603_cast_fp16)[name = string("reshape_904_cast_fp16")]; bool matmul_301_transpose_x_0 = const()[name = string("matmul_301_transpose_x_0"), val = bool(false)]; bool matmul_301_transpose_y_0 = const()[name = string("matmul_301_transpose_y_0"), val = bool(false)]; tensor matmul_301_cast_fp16 = matmul(transpose_x = matmul_301_transpose_x_0, transpose_y = matmul_301_transpose_y_0, x = reshape_903_cast_fp16, y = reshape_904_cast_fp16)[name = string("matmul_301_cast_fp16")]; tensor concat_3019 = const()[name = string("concat_3019"), val = tensor([1, 1, 104, 104])]; tensor reshape_905_cast_fp16 = reshape(shape = concat_3019, x = matmul_301_cast_fp16)[name = string("reshape_905_cast_fp16")]; tensor transpose_2989_perm_0 = const()[name = string("transpose_2989_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2989 = transpose(perm = transpose_2989_perm_0, x = reshape_905_cast_fp16)[name = string("transpose_3318")]; tensor w_1207_cast_fp16 = add(x = transpose_2989, y = transpose_2305)[name = string("w_1207_cast_fp16")]; tensor var_5775_cast_fp16 = softmax(axis = var_5559, x = w_1207_cast_fp16)[name = string("op_5775_cast_fp16")]; string var_5777_equation_0 = const()[name = string("op_5777_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5777_cast_fp16 = einsum(equation = var_5777_equation_0, values = (var_5649_cast_fp16_13, var_5775_cast_fp16))[name = string("op_5777_cast_fp16")]; tensor transpose_604_perm_0 = const()[name = string("transpose_604_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3024 = const()[name = string("concat_3024"), val = tensor([1, 104, 64])]; tensor transpose_604_cast_fp16 = transpose(perm = transpose_604_perm_0, x = var_5615_cast_fp16_14)[name = string("transpose_3317")]; tensor reshape_906_cast_fp16 = reshape(shape = concat_3024, x = transpose_604_cast_fp16)[name = string("reshape_906_cast_fp16")]; tensor transpose_605_perm_0 = const()[name = string("transpose_605_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3025 = const()[name = string("concat_3025"), val = tensor([1, 64, 104])]; tensor transpose_605_cast_fp16 = transpose(perm = transpose_605_perm_0, x = var_5632_cast_fp16_14)[name = string("transpose_3316")]; tensor reshape_907_cast_fp16 = reshape(shape = concat_3025, x = transpose_605_cast_fp16)[name = string("reshape_907_cast_fp16")]; bool matmul_302_transpose_x_0 = const()[name = string("matmul_302_transpose_x_0"), val = bool(false)]; bool matmul_302_transpose_y_0 = const()[name = string("matmul_302_transpose_y_0"), val = bool(false)]; tensor matmul_302_cast_fp16 = matmul(transpose_x = matmul_302_transpose_x_0, transpose_y = matmul_302_transpose_y_0, x = reshape_906_cast_fp16, y = reshape_907_cast_fp16)[name = string("matmul_302_cast_fp16")]; tensor concat_3029 = const()[name = string("concat_3029"), val = tensor([1, 1, 104, 104])]; tensor reshape_908_cast_fp16 = reshape(shape = concat_3029, x = matmul_302_cast_fp16)[name = string("reshape_908_cast_fp16")]; tensor transpose_2990_perm_0 = const()[name = string("transpose_2990_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2990 = transpose(perm = transpose_2990_perm_0, x = reshape_908_cast_fp16)[name = string("transpose_3315")]; tensor w_1211_cast_fp16 = add(x = transpose_2990, y = transpose_2305)[name = string("w_1211_cast_fp16")]; tensor var_5783_cast_fp16 = softmax(axis = var_5559, x = w_1211_cast_fp16)[name = string("op_5783_cast_fp16")]; string var_5785_equation_0 = const()[name = string("op_5785_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5785_cast_fp16 = einsum(equation = var_5785_equation_0, values = (var_5649_cast_fp16_14, var_5783_cast_fp16))[name = string("op_5785_cast_fp16")]; tensor transpose_606_perm_0 = const()[name = string("transpose_606_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3034 = const()[name = string("concat_3034"), val = tensor([1, 104, 64])]; tensor transpose_606_cast_fp16 = transpose(perm = transpose_606_perm_0, x = var_5615_cast_fp16_15)[name = string("transpose_3314")]; tensor reshape_909_cast_fp16 = reshape(shape = concat_3034, x = transpose_606_cast_fp16)[name = string("reshape_909_cast_fp16")]; tensor transpose_607_perm_0 = const()[name = string("transpose_607_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3035 = const()[name = string("concat_3035"), val = tensor([1, 64, 104])]; tensor transpose_607_cast_fp16 = transpose(perm = transpose_607_perm_0, x = var_5632_cast_fp16_15)[name = string("transpose_3313")]; tensor reshape_910_cast_fp16 = reshape(shape = concat_3035, x = transpose_607_cast_fp16)[name = string("reshape_910_cast_fp16")]; bool matmul_303_transpose_x_0 = const()[name = string("matmul_303_transpose_x_0"), val = bool(false)]; bool matmul_303_transpose_y_0 = const()[name = string("matmul_303_transpose_y_0"), val = bool(false)]; tensor matmul_303_cast_fp16 = matmul(transpose_x = matmul_303_transpose_x_0, transpose_y = matmul_303_transpose_y_0, x = reshape_909_cast_fp16, y = reshape_910_cast_fp16)[name = string("matmul_303_cast_fp16")]; tensor concat_3039 = const()[name = string("concat_3039"), val = tensor([1, 1, 104, 104])]; tensor reshape_911_cast_fp16 = reshape(shape = concat_3039, x = matmul_303_cast_fp16)[name = string("reshape_911_cast_fp16")]; tensor transpose_2991_perm_0 = const()[name = string("transpose_2991_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2991 = transpose(perm = transpose_2991_perm_0, x = reshape_911_cast_fp16)[name = string("transpose_3312")]; tensor w_1215_cast_fp16 = add(x = transpose_2991, y = transpose_2305)[name = string("w_1215_cast_fp16")]; tensor var_5791_cast_fp16 = softmax(axis = var_5559, x = w_1215_cast_fp16)[name = string("op_5791_cast_fp16")]; string var_5793_equation_0 = const()[name = string("op_5793_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5793_cast_fp16 = einsum(equation = var_5793_equation_0, values = (var_5649_cast_fp16_15, var_5791_cast_fp16))[name = string("op_5793_cast_fp16")]; bool input_155_interleave_0 = const()[name = string("input_155_interleave_0"), val = bool(false)]; tensor input_155_cast_fp16 = concat(axis = var_5559, interleave = input_155_interleave_0, values = (var_5673_cast_fp16, var_5681_cast_fp16, var_5689_cast_fp16, var_5697_cast_fp16, var_5705_cast_fp16, var_5713_cast_fp16, var_5721_cast_fp16, var_5729_cast_fp16, var_5737_cast_fp16, var_5745_cast_fp16, var_5753_cast_fp16, var_5761_cast_fp16, var_5769_cast_fp16, var_5777_cast_fp16, var_5785_cast_fp16, var_5793_cast_fp16))[name = string("input_155_cast_fp16")]; string var_5802_pad_type_0 = const()[name = string("op_5802_pad_type_0"), val = string("valid")]; tensor var_5802_strides_0 = const()[name = string("op_5802_strides_0"), val = tensor([1, 1])]; tensor var_5802_pad_0 = const()[name = string("op_5802_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5802_dilations_0 = const()[name = string("op_5802_dilations_0"), val = tensor([1, 1])]; int32 var_5802_groups_0 = const()[name = string("op_5802_groups_0"), val = int32(1)]; tensor layers_18_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_18_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(483857984)))]; tensor layers_18_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_18_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(485955200)))]; tensor var_5802_cast_fp16 = conv(bias = layers_18_self_attn_out_proj_bias_to_fp16, dilations = var_5802_dilations_0, groups = var_5802_groups_0, pad = var_5802_pad_0, pad_type = var_5802_pad_type_0, strides = var_5802_strides_0, weight = layers_18_self_attn_out_proj_weight_to_fp16, x = input_155_cast_fp16)[name = string("op_5802_cast_fp16")]; tensor x_197_cast_fp16 = add(x = x_193_cast_fp16, y = var_5802_cast_fp16)[name = string("x_197_cast_fp16")]; tensor mu_75_axes_0 = const()[name = string("mu_75_axes_0"), val = tensor([1])]; bool mu_75_keep_dims_0 = const()[name = string("mu_75_keep_dims_0"), val = bool(true)]; tensor mu_75_cast_fp16 = reduce_mean(axes = mu_75_axes_0, keep_dims = mu_75_keep_dims_0, x = x_197_cast_fp16)[name = string("mu_75_cast_fp16")]; tensor var_5808_cast_fp16 = sub(x = x_197_cast_fp16, y = mu_75_cast_fp16)[name = string("op_5808_cast_fp16")]; fp16 var_5562_promoted_1_to_fp16 = const()[name = string("op_5562_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_5809_cast_fp16 = pow(x = var_5808_cast_fp16, y = var_5562_promoted_1_to_fp16)[name = string("op_5809_cast_fp16")]; tensor var_75_axes_0 = const()[name = string("var_75_axes_0"), val = tensor([1])]; bool var_75_keep_dims_0 = const()[name = string("var_75_keep_dims_0"), val = bool(true)]; tensor var_75_cast_fp16 = reduce_mean(axes = var_75_axes_0, keep_dims = var_75_keep_dims_0, x = var_5809_cast_fp16)[name = string("var_75_cast_fp16")]; fp16 var_5813_to_fp16 = const()[name = string("op_5813_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_5814_cast_fp16 = add(x = var_75_cast_fp16, y = var_5813_to_fp16)[name = string("op_5814_cast_fp16")]; fp32 var_5815_epsilon_0 = const()[name = string("op_5815_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_5815_cast_fp16 = rsqrt(epsilon = var_5815_epsilon_0, x = var_5814_cast_fp16)[name = string("op_5815_cast_fp16")]; tensor x_199_cast_fp16 = mul(x = var_5808_cast_fp16, y = var_5815_cast_fp16)[name = string("x_199_cast_fp16")]; tensor input_157_gamma_0_to_fp16 = const()[name = string("input_157_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(485957312)))]; tensor input_157_beta_0_to_fp16 = const()[name = string("input_157_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(485959424)))]; fp16 input_157_epsilon_0_to_fp16 = const()[name = string("input_157_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_157_cast_fp16 = batch_norm(beta = input_157_beta_0_to_fp16, epsilon = input_157_epsilon_0_to_fp16, gamma = input_157_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_199_cast_fp16)[name = string("input_157_cast_fp16")]; string x_201_pad_type_0 = const()[name = string("x_201_pad_type_0"), val = string("valid")]; tensor x_201_strides_0 = const()[name = string("x_201_strides_0"), val = tensor([1, 1])]; tensor x_201_pad_0 = const()[name = string("x_201_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_201_dilations_0 = const()[name = string("x_201_dilations_0"), val = tensor([1, 1])]; int32 x_201_groups_0 = const()[name = string("x_201_groups_0"), val = int32(1)]; tensor layers_18_fc1_weight_to_fp16 = const()[name = string("layers_18_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(485961536)))]; tensor layers_18_fc1_bias_to_fp16 = const()[name = string("layers_18_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(494350208)))]; tensor x_201_cast_fp16 = conv(bias = layers_18_fc1_bias_to_fp16, dilations = x_201_dilations_0, groups = x_201_groups_0, pad = x_201_pad_0, pad_type = x_201_pad_type_0, strides = x_201_strides_0, weight = layers_18_fc1_weight_to_fp16, x = input_157_cast_fp16)[name = string("x_201_cast_fp16")]; fp16 var_5830_to_fp16 = const()[name = string("op_5830_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_5831_cast_fp16 = mul(x = x_201_cast_fp16, y = var_5830_to_fp16)[name = string("op_5831_cast_fp16")]; tensor var_5832_cast_fp16 = mul(x = var_5831_cast_fp16, y = x_201_cast_fp16)[name = string("op_5832_cast_fp16")]; tensor var_5833_cast_fp16 = mul(x = var_5832_cast_fp16, y = x_201_cast_fp16)[name = string("op_5833_cast_fp16")]; tensor var_5834_cast_fp16 = add(x = x_201_cast_fp16, y = var_5833_cast_fp16)[name = string("op_5834_cast_fp16")]; fp16 var_5835_to_fp16 = const()[name = string("op_5835_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_43_cast_fp16 = mul(x = var_5834_cast_fp16, y = var_5835_to_fp16)[name = string("u_43_cast_fp16")]; fp16 var_5837_to_fp16 = const()[name = string("op_5837_to_fp16"), val = fp16(0x1p-1)]; tensor var_5838_cast_fp16 = mul(x = x_201_cast_fp16, y = var_5837_to_fp16)[name = string("op_5838_cast_fp16")]; tensor var_5839_cast_fp16 = tanh(x = u_43_cast_fp16)[name = string("op_5839_cast_fp16")]; fp16 var_5840_to_fp16 = const()[name = string("op_5840_to_fp16"), val = fp16(0x1p+0)]; tensor var_5841_cast_fp16 = add(x = var_5839_cast_fp16, y = var_5840_to_fp16)[name = string("op_5841_cast_fp16")]; tensor input_159_cast_fp16 = mul(x = var_5838_cast_fp16, y = var_5841_cast_fp16)[name = string("input_159_cast_fp16")]; string h_37_pad_type_0 = const()[name = string("h_37_pad_type_0"), val = string("valid")]; tensor h_37_strides_0 = const()[name = string("h_37_strides_0"), val = tensor([1, 1])]; tensor h_37_pad_0 = const()[name = string("h_37_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_37_dilations_0 = const()[name = string("h_37_dilations_0"), val = tensor([1, 1])]; int32 h_37_groups_0 = const()[name = string("h_37_groups_0"), val = int32(1)]; tensor layers_18_fc2_weight_to_fp16 = const()[name = string("layers_18_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(494358464)))]; tensor layers_18_fc2_bias_to_fp16 = const()[name = string("layers_18_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(502747136)))]; tensor h_37_cast_fp16 = conv(bias = layers_18_fc2_bias_to_fp16, dilations = h_37_dilations_0, groups = h_37_groups_0, pad = h_37_pad_0, pad_type = h_37_pad_type_0, strides = h_37_strides_0, weight = layers_18_fc2_weight_to_fp16, x = input_159_cast_fp16)[name = string("h_37_cast_fp16")]; tensor x_203_cast_fp16 = add(x = x_197_cast_fp16, y = h_37_cast_fp16)[name = string("x_203_cast_fp16")]; int32 var_5857 = const()[name = string("op_5857"), val = int32(1)]; tensor mu_77_axes_0 = const()[name = string("mu_77_axes_0"), val = tensor([1])]; bool mu_77_keep_dims_0 = const()[name = string("mu_77_keep_dims_0"), val = bool(true)]; tensor mu_77_cast_fp16 = reduce_mean(axes = mu_77_axes_0, keep_dims = mu_77_keep_dims_0, x = x_203_cast_fp16)[name = string("mu_77_cast_fp16")]; tensor var_5871_cast_fp16 = sub(x = x_203_cast_fp16, y = mu_77_cast_fp16)[name = string("op_5871_cast_fp16")]; fp16 var_5860_promoted_to_fp16 = const()[name = string("op_5860_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_5872_cast_fp16 = pow(x = var_5871_cast_fp16, y = var_5860_promoted_to_fp16)[name = string("op_5872_cast_fp16")]; tensor var_77_axes_0 = const()[name = string("var_77_axes_0"), val = tensor([1])]; bool var_77_keep_dims_0 = const()[name = string("var_77_keep_dims_0"), val = bool(true)]; tensor var_77_cast_fp16 = reduce_mean(axes = var_77_axes_0, keep_dims = var_77_keep_dims_0, x = var_5872_cast_fp16)[name = string("var_77_cast_fp16")]; fp16 var_5876_to_fp16 = const()[name = string("op_5876_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_5877_cast_fp16 = add(x = var_77_cast_fp16, y = var_5876_to_fp16)[name = string("op_5877_cast_fp16")]; fp32 var_5878_epsilon_0 = const()[name = string("op_5878_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_5878_cast_fp16 = rsqrt(epsilon = var_5878_epsilon_0, x = var_5877_cast_fp16)[name = string("op_5878_cast_fp16")]; tensor x_205_cast_fp16 = mul(x = var_5871_cast_fp16, y = var_5878_cast_fp16)[name = string("x_205_cast_fp16")]; tensor input_161_gamma_0_to_fp16 = const()[name = string("input_161_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(502749248)))]; tensor input_161_beta_0_to_fp16 = const()[name = string("input_161_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(502751360)))]; fp16 input_161_epsilon_0_to_fp16 = const()[name = string("input_161_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_161_cast_fp16 = batch_norm(beta = input_161_beta_0_to_fp16, epsilon = input_161_epsilon_0_to_fp16, gamma = input_161_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_205_cast_fp16)[name = string("input_161_cast_fp16")]; string var_5896_pad_type_0 = const()[name = string("op_5896_pad_type_0"), val = string("valid")]; tensor var_5896_strides_0 = const()[name = string("op_5896_strides_0"), val = tensor([1, 1])]; tensor var_5896_pad_0 = const()[name = string("op_5896_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5896_dilations_0 = const()[name = string("op_5896_dilations_0"), val = tensor([1, 1])]; int32 var_5896_groups_0 = const()[name = string("op_5896_groups_0"), val = int32(1)]; tensor var_5898_weight_0_to_fp16 = const()[name = string("op_5898_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(502753472)))]; tensor var_5898_bias_0_to_fp16 = const()[name = string("op_5898_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(504850688)))]; tensor var_5898_cast_fp16 = conv(bias = var_5898_bias_0_to_fp16, dilations = var_5896_dilations_0, groups = var_5896_groups_0, pad = var_5896_pad_0, pad_type = var_5896_pad_type_0, strides = var_5896_strides_0, weight = var_5898_weight_0_to_fp16, x = input_161_cast_fp16)[name = string("op_5898_cast_fp16")]; string var_5905_pad_type_0 = const()[name = string("op_5905_pad_type_0"), val = string("valid")]; tensor var_5905_strides_0 = const()[name = string("op_5905_strides_0"), val = tensor([1, 1])]; tensor var_5905_pad_0 = const()[name = string("op_5905_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5905_dilations_0 = const()[name = string("op_5905_dilations_0"), val = tensor([1, 1])]; int32 var_5905_groups_0 = const()[name = string("op_5905_groups_0"), val = int32(1)]; tensor layers_19_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_19_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(504852800)))]; tensor layers_19_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_19_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(506950016)))]; tensor var_5905_cast_fp16 = conv(bias = layers_19_self_attn_k_proj_bias_to_fp16, dilations = var_5905_dilations_0, groups = var_5905_groups_0, pad = var_5905_pad_0, pad_type = var_5905_pad_type_0, strides = var_5905_strides_0, weight = layers_19_self_attn_k_proj_weight_to_fp16, x = input_161_cast_fp16)[name = string("op_5905_cast_fp16")]; string var_5912_pad_type_0 = const()[name = string("op_5912_pad_type_0"), val = string("valid")]; tensor var_5912_strides_0 = const()[name = string("op_5912_strides_0"), val = tensor([1, 1])]; tensor var_5912_pad_0 = const()[name = string("op_5912_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_5912_dilations_0 = const()[name = string("op_5912_dilations_0"), val = tensor([1, 1])]; int32 var_5912_groups_0 = const()[name = string("op_5912_groups_0"), val = int32(1)]; tensor layers_19_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_19_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(506952128)))]; tensor layers_19_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_19_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(509049344)))]; tensor var_5912_cast_fp16 = conv(bias = layers_19_self_attn_v_proj_bias_to_fp16, dilations = var_5912_dilations_0, groups = var_5912_groups_0, pad = var_5912_pad_0, pad_type = var_5912_pad_type_0, strides = var_5912_strides_0, weight = layers_19_self_attn_v_proj_weight_to_fp16, x = input_161_cast_fp16)[name = string("op_5912_cast_fp16")]; tensor tile_57 = const()[name = string("tile_57"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(509051456)))]; int32 var_5913_axis_0 = const()[name = string("op_5913_axis_0"), val = int32(1)]; tensor var_5913_cast_fp16_0, tensor var_5913_cast_fp16_1, tensor var_5913_cast_fp16_2, tensor var_5913_cast_fp16_3, tensor var_5913_cast_fp16_4, tensor var_5913_cast_fp16_5, tensor var_5913_cast_fp16_6, tensor var_5913_cast_fp16_7, tensor var_5913_cast_fp16_8, tensor var_5913_cast_fp16_9, tensor var_5913_cast_fp16_10, tensor var_5913_cast_fp16_11, tensor var_5913_cast_fp16_12, tensor var_5913_cast_fp16_13, tensor var_5913_cast_fp16_14, tensor var_5913_cast_fp16_15 = split(axis = var_5913_axis_0, split_sizes = tile_57, x = var_5898_cast_fp16)[name = string("op_5913_cast_fp16")]; tensor tile_58 = const()[name = string("tile_58"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(509051584)))]; int32 var_5930_axis_0 = const()[name = string("op_5930_axis_0"), val = int32(1)]; tensor var_5930_cast_fp16_0, tensor var_5930_cast_fp16_1, tensor var_5930_cast_fp16_2, tensor var_5930_cast_fp16_3, tensor var_5930_cast_fp16_4, tensor var_5930_cast_fp16_5, tensor var_5930_cast_fp16_6, tensor var_5930_cast_fp16_7, tensor var_5930_cast_fp16_8, tensor var_5930_cast_fp16_9, tensor var_5930_cast_fp16_10, tensor var_5930_cast_fp16_11, tensor var_5930_cast_fp16_12, tensor var_5930_cast_fp16_13, tensor var_5930_cast_fp16_14, tensor var_5930_cast_fp16_15 = split(axis = var_5930_axis_0, split_sizes = tile_58, x = var_5905_cast_fp16)[name = string("op_5930_cast_fp16")]; tensor tile_59 = const()[name = string("tile_59"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(509051712)))]; int32 var_5947_axis_0 = const()[name = string("op_5947_axis_0"), val = int32(1)]; tensor var_5947_cast_fp16_0, tensor var_5947_cast_fp16_1, tensor var_5947_cast_fp16_2, tensor var_5947_cast_fp16_3, tensor var_5947_cast_fp16_4, tensor var_5947_cast_fp16_5, tensor var_5947_cast_fp16_6, tensor var_5947_cast_fp16_7, tensor var_5947_cast_fp16_8, tensor var_5947_cast_fp16_9, tensor var_5947_cast_fp16_10, tensor var_5947_cast_fp16_11, tensor var_5947_cast_fp16_12, tensor var_5947_cast_fp16_13, tensor var_5947_cast_fp16_14, tensor var_5947_cast_fp16_15 = split(axis = var_5947_axis_0, split_sizes = tile_59, x = var_5912_cast_fp16)[name = string("op_5947_cast_fp16")]; tensor transpose_608_perm_0 = const()[name = string("transpose_608_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3044 = const()[name = string("concat_3044"), val = tensor([1, 104, 64])]; tensor transpose_608_cast_fp16 = transpose(perm = transpose_608_perm_0, x = var_5913_cast_fp16_0)[name = string("transpose_3311")]; tensor reshape_912_cast_fp16 = reshape(shape = concat_3044, x = transpose_608_cast_fp16)[name = string("reshape_912_cast_fp16")]; tensor transpose_609_perm_0 = const()[name = string("transpose_609_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3045 = const()[name = string("concat_3045"), val = tensor([1, 64, 104])]; tensor transpose_609_cast_fp16 = transpose(perm = transpose_609_perm_0, x = var_5930_cast_fp16_0)[name = string("transpose_3310")]; tensor reshape_913_cast_fp16 = reshape(shape = concat_3045, x = transpose_609_cast_fp16)[name = string("reshape_913_cast_fp16")]; bool matmul_304_transpose_x_0 = const()[name = string("matmul_304_transpose_x_0"), val = bool(false)]; bool matmul_304_transpose_y_0 = const()[name = string("matmul_304_transpose_y_0"), val = bool(false)]; tensor matmul_304_cast_fp16 = matmul(transpose_x = matmul_304_transpose_x_0, transpose_y = matmul_304_transpose_y_0, x = reshape_912_cast_fp16, y = reshape_913_cast_fp16)[name = string("matmul_304_cast_fp16")]; tensor concat_3049 = const()[name = string("concat_3049"), val = tensor([1, 1, 104, 104])]; tensor reshape_914_cast_fp16 = reshape(shape = concat_3049, x = matmul_304_cast_fp16)[name = string("reshape_914_cast_fp16")]; tensor transpose_2992_perm_0 = const()[name = string("transpose_2992_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2992 = transpose(perm = transpose_2992_perm_0, x = reshape_914_cast_fp16)[name = string("transpose_3309")]; tensor w_1219_cast_fp16 = add(x = transpose_2992, y = transpose_2305)[name = string("w_1219_cast_fp16")]; tensor var_5969_cast_fp16 = softmax(axis = var_5857, x = w_1219_cast_fp16)[name = string("op_5969_cast_fp16")]; string var_5971_equation_0 = const()[name = string("op_5971_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5971_cast_fp16 = einsum(equation = var_5971_equation_0, values = (var_5947_cast_fp16_0, var_5969_cast_fp16))[name = string("op_5971_cast_fp16")]; tensor transpose_610_perm_0 = const()[name = string("transpose_610_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3054 = const()[name = string("concat_3054"), val = tensor([1, 104, 64])]; tensor transpose_610_cast_fp16 = transpose(perm = transpose_610_perm_0, x = var_5913_cast_fp16_1)[name = string("transpose_3308")]; tensor reshape_915_cast_fp16 = reshape(shape = concat_3054, x = transpose_610_cast_fp16)[name = string("reshape_915_cast_fp16")]; tensor transpose_611_perm_0 = const()[name = string("transpose_611_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3055 = const()[name = string("concat_3055"), val = tensor([1, 64, 104])]; tensor transpose_611_cast_fp16 = transpose(perm = transpose_611_perm_0, x = var_5930_cast_fp16_1)[name = string("transpose_3307")]; tensor reshape_916_cast_fp16 = reshape(shape = concat_3055, x = transpose_611_cast_fp16)[name = string("reshape_916_cast_fp16")]; bool matmul_305_transpose_x_0 = const()[name = string("matmul_305_transpose_x_0"), val = bool(false)]; bool matmul_305_transpose_y_0 = const()[name = string("matmul_305_transpose_y_0"), val = bool(false)]; tensor matmul_305_cast_fp16 = matmul(transpose_x = matmul_305_transpose_x_0, transpose_y = matmul_305_transpose_y_0, x = reshape_915_cast_fp16, y = reshape_916_cast_fp16)[name = string("matmul_305_cast_fp16")]; tensor concat_3059 = const()[name = string("concat_3059"), val = tensor([1, 1, 104, 104])]; tensor reshape_917_cast_fp16 = reshape(shape = concat_3059, x = matmul_305_cast_fp16)[name = string("reshape_917_cast_fp16")]; tensor transpose_2993_perm_0 = const()[name = string("transpose_2993_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2993 = transpose(perm = transpose_2993_perm_0, x = reshape_917_cast_fp16)[name = string("transpose_3306")]; tensor w_1223_cast_fp16 = add(x = transpose_2993, y = transpose_2305)[name = string("w_1223_cast_fp16")]; tensor var_5977_cast_fp16 = softmax(axis = var_5857, x = w_1223_cast_fp16)[name = string("op_5977_cast_fp16")]; string var_5979_equation_0 = const()[name = string("op_5979_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5979_cast_fp16 = einsum(equation = var_5979_equation_0, values = (var_5947_cast_fp16_1, var_5977_cast_fp16))[name = string("op_5979_cast_fp16")]; tensor transpose_612_perm_0 = const()[name = string("transpose_612_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3064 = const()[name = string("concat_3064"), val = tensor([1, 104, 64])]; tensor transpose_612_cast_fp16 = transpose(perm = transpose_612_perm_0, x = var_5913_cast_fp16_2)[name = string("transpose_3305")]; tensor reshape_918_cast_fp16 = reshape(shape = concat_3064, x = transpose_612_cast_fp16)[name = string("reshape_918_cast_fp16")]; tensor transpose_613_perm_0 = const()[name = string("transpose_613_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3065 = const()[name = string("concat_3065"), val = tensor([1, 64, 104])]; tensor transpose_613_cast_fp16 = transpose(perm = transpose_613_perm_0, x = var_5930_cast_fp16_2)[name = string("transpose_3304")]; tensor reshape_919_cast_fp16 = reshape(shape = concat_3065, x = transpose_613_cast_fp16)[name = string("reshape_919_cast_fp16")]; bool matmul_306_transpose_x_0 = const()[name = string("matmul_306_transpose_x_0"), val = bool(false)]; bool matmul_306_transpose_y_0 = const()[name = string("matmul_306_transpose_y_0"), val = bool(false)]; tensor matmul_306_cast_fp16 = matmul(transpose_x = matmul_306_transpose_x_0, transpose_y = matmul_306_transpose_y_0, x = reshape_918_cast_fp16, y = reshape_919_cast_fp16)[name = string("matmul_306_cast_fp16")]; tensor concat_3069 = const()[name = string("concat_3069"), val = tensor([1, 1, 104, 104])]; tensor reshape_920_cast_fp16 = reshape(shape = concat_3069, x = matmul_306_cast_fp16)[name = string("reshape_920_cast_fp16")]; tensor transpose_2994_perm_0 = const()[name = string("transpose_2994_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2994 = transpose(perm = transpose_2994_perm_0, x = reshape_920_cast_fp16)[name = string("transpose_3303")]; tensor w_1227_cast_fp16 = add(x = transpose_2994, y = transpose_2305)[name = string("w_1227_cast_fp16")]; tensor var_5985_cast_fp16 = softmax(axis = var_5857, x = w_1227_cast_fp16)[name = string("op_5985_cast_fp16")]; string var_5987_equation_0 = const()[name = string("op_5987_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5987_cast_fp16 = einsum(equation = var_5987_equation_0, values = (var_5947_cast_fp16_2, var_5985_cast_fp16))[name = string("op_5987_cast_fp16")]; tensor transpose_614_perm_0 = const()[name = string("transpose_614_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3074 = const()[name = string("concat_3074"), val = tensor([1, 104, 64])]; tensor transpose_614_cast_fp16 = transpose(perm = transpose_614_perm_0, x = var_5913_cast_fp16_3)[name = string("transpose_3302")]; tensor reshape_921_cast_fp16 = reshape(shape = concat_3074, x = transpose_614_cast_fp16)[name = string("reshape_921_cast_fp16")]; tensor transpose_615_perm_0 = const()[name = string("transpose_615_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3075 = const()[name = string("concat_3075"), val = tensor([1, 64, 104])]; tensor transpose_615_cast_fp16 = transpose(perm = transpose_615_perm_0, x = var_5930_cast_fp16_3)[name = string("transpose_3301")]; tensor reshape_922_cast_fp16 = reshape(shape = concat_3075, x = transpose_615_cast_fp16)[name = string("reshape_922_cast_fp16")]; bool matmul_307_transpose_x_0 = const()[name = string("matmul_307_transpose_x_0"), val = bool(false)]; bool matmul_307_transpose_y_0 = const()[name = string("matmul_307_transpose_y_0"), val = bool(false)]; tensor matmul_307_cast_fp16 = matmul(transpose_x = matmul_307_transpose_x_0, transpose_y = matmul_307_transpose_y_0, x = reshape_921_cast_fp16, y = reshape_922_cast_fp16)[name = string("matmul_307_cast_fp16")]; tensor concat_3079 = const()[name = string("concat_3079"), val = tensor([1, 1, 104, 104])]; tensor reshape_923_cast_fp16 = reshape(shape = concat_3079, x = matmul_307_cast_fp16)[name = string("reshape_923_cast_fp16")]; tensor transpose_2995_perm_0 = const()[name = string("transpose_2995_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2995 = transpose(perm = transpose_2995_perm_0, x = reshape_923_cast_fp16)[name = string("transpose_3300")]; tensor w_1231_cast_fp16 = add(x = transpose_2995, y = transpose_2305)[name = string("w_1231_cast_fp16")]; tensor var_5993_cast_fp16 = softmax(axis = var_5857, x = w_1231_cast_fp16)[name = string("op_5993_cast_fp16")]; string var_5995_equation_0 = const()[name = string("op_5995_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_5995_cast_fp16 = einsum(equation = var_5995_equation_0, values = (var_5947_cast_fp16_3, var_5993_cast_fp16))[name = string("op_5995_cast_fp16")]; tensor transpose_616_perm_0 = const()[name = string("transpose_616_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3084 = const()[name = string("concat_3084"), val = tensor([1, 104, 64])]; tensor transpose_616_cast_fp16 = transpose(perm = transpose_616_perm_0, x = var_5913_cast_fp16_4)[name = string("transpose_3299")]; tensor reshape_924_cast_fp16 = reshape(shape = concat_3084, x = transpose_616_cast_fp16)[name = string("reshape_924_cast_fp16")]; tensor transpose_617_perm_0 = const()[name = string("transpose_617_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3085 = const()[name = string("concat_3085"), val = tensor([1, 64, 104])]; tensor transpose_617_cast_fp16 = transpose(perm = transpose_617_perm_0, x = var_5930_cast_fp16_4)[name = string("transpose_3298")]; tensor reshape_925_cast_fp16 = reshape(shape = concat_3085, x = transpose_617_cast_fp16)[name = string("reshape_925_cast_fp16")]; bool matmul_308_transpose_x_0 = const()[name = string("matmul_308_transpose_x_0"), val = bool(false)]; bool matmul_308_transpose_y_0 = const()[name = string("matmul_308_transpose_y_0"), val = bool(false)]; tensor matmul_308_cast_fp16 = matmul(transpose_x = matmul_308_transpose_x_0, transpose_y = matmul_308_transpose_y_0, x = reshape_924_cast_fp16, y = reshape_925_cast_fp16)[name = string("matmul_308_cast_fp16")]; tensor concat_3089 = const()[name = string("concat_3089"), val = tensor([1, 1, 104, 104])]; tensor reshape_926_cast_fp16 = reshape(shape = concat_3089, x = matmul_308_cast_fp16)[name = string("reshape_926_cast_fp16")]; tensor transpose_2996_perm_0 = const()[name = string("transpose_2996_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2996 = transpose(perm = transpose_2996_perm_0, x = reshape_926_cast_fp16)[name = string("transpose_3297")]; tensor w_1235_cast_fp16 = add(x = transpose_2996, y = transpose_2305)[name = string("w_1235_cast_fp16")]; tensor var_6001_cast_fp16 = softmax(axis = var_5857, x = w_1235_cast_fp16)[name = string("op_6001_cast_fp16")]; string var_6003_equation_0 = const()[name = string("op_6003_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6003_cast_fp16 = einsum(equation = var_6003_equation_0, values = (var_5947_cast_fp16_4, var_6001_cast_fp16))[name = string("op_6003_cast_fp16")]; tensor transpose_618_perm_0 = const()[name = string("transpose_618_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3094 = const()[name = string("concat_3094"), val = tensor([1, 104, 64])]; tensor transpose_618_cast_fp16 = transpose(perm = transpose_618_perm_0, x = var_5913_cast_fp16_5)[name = string("transpose_3296")]; tensor reshape_927_cast_fp16 = reshape(shape = concat_3094, x = transpose_618_cast_fp16)[name = string("reshape_927_cast_fp16")]; tensor transpose_619_perm_0 = const()[name = string("transpose_619_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3095 = const()[name = string("concat_3095"), val = tensor([1, 64, 104])]; tensor transpose_619_cast_fp16 = transpose(perm = transpose_619_perm_0, x = var_5930_cast_fp16_5)[name = string("transpose_3295")]; tensor reshape_928_cast_fp16 = reshape(shape = concat_3095, x = transpose_619_cast_fp16)[name = string("reshape_928_cast_fp16")]; bool matmul_309_transpose_x_0 = const()[name = string("matmul_309_transpose_x_0"), val = bool(false)]; bool matmul_309_transpose_y_0 = const()[name = string("matmul_309_transpose_y_0"), val = bool(false)]; tensor matmul_309_cast_fp16 = matmul(transpose_x = matmul_309_transpose_x_0, transpose_y = matmul_309_transpose_y_0, x = reshape_927_cast_fp16, y = reshape_928_cast_fp16)[name = string("matmul_309_cast_fp16")]; tensor concat_3099 = const()[name = string("concat_3099"), val = tensor([1, 1, 104, 104])]; tensor reshape_929_cast_fp16 = reshape(shape = concat_3099, x = matmul_309_cast_fp16)[name = string("reshape_929_cast_fp16")]; tensor transpose_2997_perm_0 = const()[name = string("transpose_2997_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2997 = transpose(perm = transpose_2997_perm_0, x = reshape_929_cast_fp16)[name = string("transpose_3294")]; tensor w_1239_cast_fp16 = add(x = transpose_2997, y = transpose_2305)[name = string("w_1239_cast_fp16")]; tensor var_6009_cast_fp16 = softmax(axis = var_5857, x = w_1239_cast_fp16)[name = string("op_6009_cast_fp16")]; string var_6011_equation_0 = const()[name = string("op_6011_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6011_cast_fp16 = einsum(equation = var_6011_equation_0, values = (var_5947_cast_fp16_5, var_6009_cast_fp16))[name = string("op_6011_cast_fp16")]; tensor transpose_620_perm_0 = const()[name = string("transpose_620_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3104 = const()[name = string("concat_3104"), val = tensor([1, 104, 64])]; tensor transpose_620_cast_fp16 = transpose(perm = transpose_620_perm_0, x = var_5913_cast_fp16_6)[name = string("transpose_3293")]; tensor reshape_930_cast_fp16 = reshape(shape = concat_3104, x = transpose_620_cast_fp16)[name = string("reshape_930_cast_fp16")]; tensor transpose_621_perm_0 = const()[name = string("transpose_621_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3105 = const()[name = string("concat_3105"), val = tensor([1, 64, 104])]; tensor transpose_621_cast_fp16 = transpose(perm = transpose_621_perm_0, x = var_5930_cast_fp16_6)[name = string("transpose_3292")]; tensor reshape_931_cast_fp16 = reshape(shape = concat_3105, x = transpose_621_cast_fp16)[name = string("reshape_931_cast_fp16")]; bool matmul_310_transpose_x_0 = const()[name = string("matmul_310_transpose_x_0"), val = bool(false)]; bool matmul_310_transpose_y_0 = const()[name = string("matmul_310_transpose_y_0"), val = bool(false)]; tensor matmul_310_cast_fp16 = matmul(transpose_x = matmul_310_transpose_x_0, transpose_y = matmul_310_transpose_y_0, x = reshape_930_cast_fp16, y = reshape_931_cast_fp16)[name = string("matmul_310_cast_fp16")]; tensor concat_3109 = const()[name = string("concat_3109"), val = tensor([1, 1, 104, 104])]; tensor reshape_932_cast_fp16 = reshape(shape = concat_3109, x = matmul_310_cast_fp16)[name = string("reshape_932_cast_fp16")]; tensor transpose_2998_perm_0 = const()[name = string("transpose_2998_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2998 = transpose(perm = transpose_2998_perm_0, x = reshape_932_cast_fp16)[name = string("transpose_3291")]; tensor w_1243_cast_fp16 = add(x = transpose_2998, y = transpose_2305)[name = string("w_1243_cast_fp16")]; tensor var_6017_cast_fp16 = softmax(axis = var_5857, x = w_1243_cast_fp16)[name = string("op_6017_cast_fp16")]; string var_6019_equation_0 = const()[name = string("op_6019_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6019_cast_fp16 = einsum(equation = var_6019_equation_0, values = (var_5947_cast_fp16_6, var_6017_cast_fp16))[name = string("op_6019_cast_fp16")]; tensor transpose_622_perm_0 = const()[name = string("transpose_622_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3114 = const()[name = string("concat_3114"), val = tensor([1, 104, 64])]; tensor transpose_622_cast_fp16 = transpose(perm = transpose_622_perm_0, x = var_5913_cast_fp16_7)[name = string("transpose_3290")]; tensor reshape_933_cast_fp16 = reshape(shape = concat_3114, x = transpose_622_cast_fp16)[name = string("reshape_933_cast_fp16")]; tensor transpose_623_perm_0 = const()[name = string("transpose_623_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3115 = const()[name = string("concat_3115"), val = tensor([1, 64, 104])]; tensor transpose_623_cast_fp16 = transpose(perm = transpose_623_perm_0, x = var_5930_cast_fp16_7)[name = string("transpose_3289")]; tensor reshape_934_cast_fp16 = reshape(shape = concat_3115, x = transpose_623_cast_fp16)[name = string("reshape_934_cast_fp16")]; bool matmul_311_transpose_x_0 = const()[name = string("matmul_311_transpose_x_0"), val = bool(false)]; bool matmul_311_transpose_y_0 = const()[name = string("matmul_311_transpose_y_0"), val = bool(false)]; tensor matmul_311_cast_fp16 = matmul(transpose_x = matmul_311_transpose_x_0, transpose_y = matmul_311_transpose_y_0, x = reshape_933_cast_fp16, y = reshape_934_cast_fp16)[name = string("matmul_311_cast_fp16")]; tensor concat_3119 = const()[name = string("concat_3119"), val = tensor([1, 1, 104, 104])]; tensor reshape_935_cast_fp16 = reshape(shape = concat_3119, x = matmul_311_cast_fp16)[name = string("reshape_935_cast_fp16")]; tensor transpose_2999_perm_0 = const()[name = string("transpose_2999_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_2999 = transpose(perm = transpose_2999_perm_0, x = reshape_935_cast_fp16)[name = string("transpose_3288")]; tensor w_1247_cast_fp16 = add(x = transpose_2999, y = transpose_2305)[name = string("w_1247_cast_fp16")]; tensor var_6025_cast_fp16 = softmax(axis = var_5857, x = w_1247_cast_fp16)[name = string("op_6025_cast_fp16")]; string var_6027_equation_0 = const()[name = string("op_6027_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6027_cast_fp16 = einsum(equation = var_6027_equation_0, values = (var_5947_cast_fp16_7, var_6025_cast_fp16))[name = string("op_6027_cast_fp16")]; tensor transpose_624_perm_0 = const()[name = string("transpose_624_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3124 = const()[name = string("concat_3124"), val = tensor([1, 104, 64])]; tensor transpose_624_cast_fp16 = transpose(perm = transpose_624_perm_0, x = var_5913_cast_fp16_8)[name = string("transpose_3287")]; tensor reshape_936_cast_fp16 = reshape(shape = concat_3124, x = transpose_624_cast_fp16)[name = string("reshape_936_cast_fp16")]; tensor transpose_625_perm_0 = const()[name = string("transpose_625_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3125 = const()[name = string("concat_3125"), val = tensor([1, 64, 104])]; tensor transpose_625_cast_fp16 = transpose(perm = transpose_625_perm_0, x = var_5930_cast_fp16_8)[name = string("transpose_3286")]; tensor reshape_937_cast_fp16 = reshape(shape = concat_3125, x = transpose_625_cast_fp16)[name = string("reshape_937_cast_fp16")]; bool matmul_312_transpose_x_0 = const()[name = string("matmul_312_transpose_x_0"), val = bool(false)]; bool matmul_312_transpose_y_0 = const()[name = string("matmul_312_transpose_y_0"), val = bool(false)]; tensor matmul_312_cast_fp16 = matmul(transpose_x = matmul_312_transpose_x_0, transpose_y = matmul_312_transpose_y_0, x = reshape_936_cast_fp16, y = reshape_937_cast_fp16)[name = string("matmul_312_cast_fp16")]; tensor concat_3129 = const()[name = string("concat_3129"), val = tensor([1, 1, 104, 104])]; tensor reshape_938_cast_fp16 = reshape(shape = concat_3129, x = matmul_312_cast_fp16)[name = string("reshape_938_cast_fp16")]; tensor transpose_3000_perm_0 = const()[name = string("transpose_3000_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3000 = transpose(perm = transpose_3000_perm_0, x = reshape_938_cast_fp16)[name = string("transpose_3285")]; tensor w_1251_cast_fp16 = add(x = transpose_3000, y = transpose_2305)[name = string("w_1251_cast_fp16")]; tensor var_6033_cast_fp16 = softmax(axis = var_5857, x = w_1251_cast_fp16)[name = string("op_6033_cast_fp16")]; string var_6035_equation_0 = const()[name = string("op_6035_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6035_cast_fp16 = einsum(equation = var_6035_equation_0, values = (var_5947_cast_fp16_8, var_6033_cast_fp16))[name = string("op_6035_cast_fp16")]; tensor transpose_626_perm_0 = const()[name = string("transpose_626_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3134 = const()[name = string("concat_3134"), val = tensor([1, 104, 64])]; tensor transpose_626_cast_fp16 = transpose(perm = transpose_626_perm_0, x = var_5913_cast_fp16_9)[name = string("transpose_3284")]; tensor reshape_939_cast_fp16 = reshape(shape = concat_3134, x = transpose_626_cast_fp16)[name = string("reshape_939_cast_fp16")]; tensor transpose_627_perm_0 = const()[name = string("transpose_627_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3135 = const()[name = string("concat_3135"), val = tensor([1, 64, 104])]; tensor transpose_627_cast_fp16 = transpose(perm = transpose_627_perm_0, x = var_5930_cast_fp16_9)[name = string("transpose_3283")]; tensor reshape_940_cast_fp16 = reshape(shape = concat_3135, x = transpose_627_cast_fp16)[name = string("reshape_940_cast_fp16")]; bool matmul_313_transpose_x_0 = const()[name = string("matmul_313_transpose_x_0"), val = bool(false)]; bool matmul_313_transpose_y_0 = const()[name = string("matmul_313_transpose_y_0"), val = bool(false)]; tensor matmul_313_cast_fp16 = matmul(transpose_x = matmul_313_transpose_x_0, transpose_y = matmul_313_transpose_y_0, x = reshape_939_cast_fp16, y = reshape_940_cast_fp16)[name = string("matmul_313_cast_fp16")]; tensor concat_3139 = const()[name = string("concat_3139"), val = tensor([1, 1, 104, 104])]; tensor reshape_941_cast_fp16 = reshape(shape = concat_3139, x = matmul_313_cast_fp16)[name = string("reshape_941_cast_fp16")]; tensor transpose_3001_perm_0 = const()[name = string("transpose_3001_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3001 = transpose(perm = transpose_3001_perm_0, x = reshape_941_cast_fp16)[name = string("transpose_3282")]; tensor w_1255_cast_fp16 = add(x = transpose_3001, y = transpose_2305)[name = string("w_1255_cast_fp16")]; tensor var_6041_cast_fp16 = softmax(axis = var_5857, x = w_1255_cast_fp16)[name = string("op_6041_cast_fp16")]; string var_6043_equation_0 = const()[name = string("op_6043_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6043_cast_fp16 = einsum(equation = var_6043_equation_0, values = (var_5947_cast_fp16_9, var_6041_cast_fp16))[name = string("op_6043_cast_fp16")]; tensor transpose_628_perm_0 = const()[name = string("transpose_628_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3144 = const()[name = string("concat_3144"), val = tensor([1, 104, 64])]; tensor transpose_628_cast_fp16 = transpose(perm = transpose_628_perm_0, x = var_5913_cast_fp16_10)[name = string("transpose_3281")]; tensor reshape_942_cast_fp16 = reshape(shape = concat_3144, x = transpose_628_cast_fp16)[name = string("reshape_942_cast_fp16")]; tensor transpose_629_perm_0 = const()[name = string("transpose_629_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3145 = const()[name = string("concat_3145"), val = tensor([1, 64, 104])]; tensor transpose_629_cast_fp16 = transpose(perm = transpose_629_perm_0, x = var_5930_cast_fp16_10)[name = string("transpose_3280")]; tensor reshape_943_cast_fp16 = reshape(shape = concat_3145, x = transpose_629_cast_fp16)[name = string("reshape_943_cast_fp16")]; bool matmul_314_transpose_x_0 = const()[name = string("matmul_314_transpose_x_0"), val = bool(false)]; bool matmul_314_transpose_y_0 = const()[name = string("matmul_314_transpose_y_0"), val = bool(false)]; tensor matmul_314_cast_fp16 = matmul(transpose_x = matmul_314_transpose_x_0, transpose_y = matmul_314_transpose_y_0, x = reshape_942_cast_fp16, y = reshape_943_cast_fp16)[name = string("matmul_314_cast_fp16")]; tensor concat_3149 = const()[name = string("concat_3149"), val = tensor([1, 1, 104, 104])]; tensor reshape_944_cast_fp16 = reshape(shape = concat_3149, x = matmul_314_cast_fp16)[name = string("reshape_944_cast_fp16")]; tensor transpose_3002_perm_0 = const()[name = string("transpose_3002_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3002 = transpose(perm = transpose_3002_perm_0, x = reshape_944_cast_fp16)[name = string("transpose_3279")]; tensor w_1259_cast_fp16 = add(x = transpose_3002, y = transpose_2305)[name = string("w_1259_cast_fp16")]; tensor var_6049_cast_fp16 = softmax(axis = var_5857, x = w_1259_cast_fp16)[name = string("op_6049_cast_fp16")]; string var_6051_equation_0 = const()[name = string("op_6051_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6051_cast_fp16 = einsum(equation = var_6051_equation_0, values = (var_5947_cast_fp16_10, var_6049_cast_fp16))[name = string("op_6051_cast_fp16")]; tensor transpose_630_perm_0 = const()[name = string("transpose_630_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3154 = const()[name = string("concat_3154"), val = tensor([1, 104, 64])]; tensor transpose_630_cast_fp16 = transpose(perm = transpose_630_perm_0, x = var_5913_cast_fp16_11)[name = string("transpose_3278")]; tensor reshape_945_cast_fp16 = reshape(shape = concat_3154, x = transpose_630_cast_fp16)[name = string("reshape_945_cast_fp16")]; tensor transpose_631_perm_0 = const()[name = string("transpose_631_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3155 = const()[name = string("concat_3155"), val = tensor([1, 64, 104])]; tensor transpose_631_cast_fp16 = transpose(perm = transpose_631_perm_0, x = var_5930_cast_fp16_11)[name = string("transpose_3277")]; tensor reshape_946_cast_fp16 = reshape(shape = concat_3155, x = transpose_631_cast_fp16)[name = string("reshape_946_cast_fp16")]; bool matmul_315_transpose_x_0 = const()[name = string("matmul_315_transpose_x_0"), val = bool(false)]; bool matmul_315_transpose_y_0 = const()[name = string("matmul_315_transpose_y_0"), val = bool(false)]; tensor matmul_315_cast_fp16 = matmul(transpose_x = matmul_315_transpose_x_0, transpose_y = matmul_315_transpose_y_0, x = reshape_945_cast_fp16, y = reshape_946_cast_fp16)[name = string("matmul_315_cast_fp16")]; tensor concat_3159 = const()[name = string("concat_3159"), val = tensor([1, 1, 104, 104])]; tensor reshape_947_cast_fp16 = reshape(shape = concat_3159, x = matmul_315_cast_fp16)[name = string("reshape_947_cast_fp16")]; tensor transpose_3003_perm_0 = const()[name = string("transpose_3003_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3003 = transpose(perm = transpose_3003_perm_0, x = reshape_947_cast_fp16)[name = string("transpose_3276")]; tensor w_1263_cast_fp16 = add(x = transpose_3003, y = transpose_2305)[name = string("w_1263_cast_fp16")]; tensor var_6057_cast_fp16 = softmax(axis = var_5857, x = w_1263_cast_fp16)[name = string("op_6057_cast_fp16")]; string var_6059_equation_0 = const()[name = string("op_6059_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6059_cast_fp16 = einsum(equation = var_6059_equation_0, values = (var_5947_cast_fp16_11, var_6057_cast_fp16))[name = string("op_6059_cast_fp16")]; tensor transpose_632_perm_0 = const()[name = string("transpose_632_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3164 = const()[name = string("concat_3164"), val = tensor([1, 104, 64])]; tensor transpose_632_cast_fp16 = transpose(perm = transpose_632_perm_0, x = var_5913_cast_fp16_12)[name = string("transpose_3275")]; tensor reshape_948_cast_fp16 = reshape(shape = concat_3164, x = transpose_632_cast_fp16)[name = string("reshape_948_cast_fp16")]; tensor transpose_633_perm_0 = const()[name = string("transpose_633_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3165 = const()[name = string("concat_3165"), val = tensor([1, 64, 104])]; tensor transpose_633_cast_fp16 = transpose(perm = transpose_633_perm_0, x = var_5930_cast_fp16_12)[name = string("transpose_3274")]; tensor reshape_949_cast_fp16 = reshape(shape = concat_3165, x = transpose_633_cast_fp16)[name = string("reshape_949_cast_fp16")]; bool matmul_316_transpose_x_0 = const()[name = string("matmul_316_transpose_x_0"), val = bool(false)]; bool matmul_316_transpose_y_0 = const()[name = string("matmul_316_transpose_y_0"), val = bool(false)]; tensor matmul_316_cast_fp16 = matmul(transpose_x = matmul_316_transpose_x_0, transpose_y = matmul_316_transpose_y_0, x = reshape_948_cast_fp16, y = reshape_949_cast_fp16)[name = string("matmul_316_cast_fp16")]; tensor concat_3169 = const()[name = string("concat_3169"), val = tensor([1, 1, 104, 104])]; tensor reshape_950_cast_fp16 = reshape(shape = concat_3169, x = matmul_316_cast_fp16)[name = string("reshape_950_cast_fp16")]; tensor transpose_3004_perm_0 = const()[name = string("transpose_3004_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3004 = transpose(perm = transpose_3004_perm_0, x = reshape_950_cast_fp16)[name = string("transpose_3273")]; tensor w_1267_cast_fp16 = add(x = transpose_3004, y = transpose_2305)[name = string("w_1267_cast_fp16")]; tensor var_6065_cast_fp16 = softmax(axis = var_5857, x = w_1267_cast_fp16)[name = string("op_6065_cast_fp16")]; string var_6067_equation_0 = const()[name = string("op_6067_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6067_cast_fp16 = einsum(equation = var_6067_equation_0, values = (var_5947_cast_fp16_12, var_6065_cast_fp16))[name = string("op_6067_cast_fp16")]; tensor transpose_634_perm_0 = const()[name = string("transpose_634_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3174 = const()[name = string("concat_3174"), val = tensor([1, 104, 64])]; tensor transpose_634_cast_fp16 = transpose(perm = transpose_634_perm_0, x = var_5913_cast_fp16_13)[name = string("transpose_3272")]; tensor reshape_951_cast_fp16 = reshape(shape = concat_3174, x = transpose_634_cast_fp16)[name = string("reshape_951_cast_fp16")]; tensor transpose_635_perm_0 = const()[name = string("transpose_635_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3175 = const()[name = string("concat_3175"), val = tensor([1, 64, 104])]; tensor transpose_635_cast_fp16 = transpose(perm = transpose_635_perm_0, x = var_5930_cast_fp16_13)[name = string("transpose_3271")]; tensor reshape_952_cast_fp16 = reshape(shape = concat_3175, x = transpose_635_cast_fp16)[name = string("reshape_952_cast_fp16")]; bool matmul_317_transpose_x_0 = const()[name = string("matmul_317_transpose_x_0"), val = bool(false)]; bool matmul_317_transpose_y_0 = const()[name = string("matmul_317_transpose_y_0"), val = bool(false)]; tensor matmul_317_cast_fp16 = matmul(transpose_x = matmul_317_transpose_x_0, transpose_y = matmul_317_transpose_y_0, x = reshape_951_cast_fp16, y = reshape_952_cast_fp16)[name = string("matmul_317_cast_fp16")]; tensor concat_3179 = const()[name = string("concat_3179"), val = tensor([1, 1, 104, 104])]; tensor reshape_953_cast_fp16 = reshape(shape = concat_3179, x = matmul_317_cast_fp16)[name = string("reshape_953_cast_fp16")]; tensor transpose_3005_perm_0 = const()[name = string("transpose_3005_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3005 = transpose(perm = transpose_3005_perm_0, x = reshape_953_cast_fp16)[name = string("transpose_3270")]; tensor w_1271_cast_fp16 = add(x = transpose_3005, y = transpose_2305)[name = string("w_1271_cast_fp16")]; tensor var_6073_cast_fp16 = softmax(axis = var_5857, x = w_1271_cast_fp16)[name = string("op_6073_cast_fp16")]; string var_6075_equation_0 = const()[name = string("op_6075_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6075_cast_fp16 = einsum(equation = var_6075_equation_0, values = (var_5947_cast_fp16_13, var_6073_cast_fp16))[name = string("op_6075_cast_fp16")]; tensor transpose_636_perm_0 = const()[name = string("transpose_636_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3184 = const()[name = string("concat_3184"), val = tensor([1, 104, 64])]; tensor transpose_636_cast_fp16 = transpose(perm = transpose_636_perm_0, x = var_5913_cast_fp16_14)[name = string("transpose_3269")]; tensor reshape_954_cast_fp16 = reshape(shape = concat_3184, x = transpose_636_cast_fp16)[name = string("reshape_954_cast_fp16")]; tensor transpose_637_perm_0 = const()[name = string("transpose_637_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3185 = const()[name = string("concat_3185"), val = tensor([1, 64, 104])]; tensor transpose_637_cast_fp16 = transpose(perm = transpose_637_perm_0, x = var_5930_cast_fp16_14)[name = string("transpose_3268")]; tensor reshape_955_cast_fp16 = reshape(shape = concat_3185, x = transpose_637_cast_fp16)[name = string("reshape_955_cast_fp16")]; bool matmul_318_transpose_x_0 = const()[name = string("matmul_318_transpose_x_0"), val = bool(false)]; bool matmul_318_transpose_y_0 = const()[name = string("matmul_318_transpose_y_0"), val = bool(false)]; tensor matmul_318_cast_fp16 = matmul(transpose_x = matmul_318_transpose_x_0, transpose_y = matmul_318_transpose_y_0, x = reshape_954_cast_fp16, y = reshape_955_cast_fp16)[name = string("matmul_318_cast_fp16")]; tensor concat_3189 = const()[name = string("concat_3189"), val = tensor([1, 1, 104, 104])]; tensor reshape_956_cast_fp16 = reshape(shape = concat_3189, x = matmul_318_cast_fp16)[name = string("reshape_956_cast_fp16")]; tensor transpose_3006_perm_0 = const()[name = string("transpose_3006_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3006 = transpose(perm = transpose_3006_perm_0, x = reshape_956_cast_fp16)[name = string("transpose_3267")]; tensor w_1275_cast_fp16 = add(x = transpose_3006, y = transpose_2305)[name = string("w_1275_cast_fp16")]; tensor var_6081_cast_fp16 = softmax(axis = var_5857, x = w_1275_cast_fp16)[name = string("op_6081_cast_fp16")]; string var_6083_equation_0 = const()[name = string("op_6083_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6083_cast_fp16 = einsum(equation = var_6083_equation_0, values = (var_5947_cast_fp16_14, var_6081_cast_fp16))[name = string("op_6083_cast_fp16")]; tensor transpose_638_perm_0 = const()[name = string("transpose_638_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3194 = const()[name = string("concat_3194"), val = tensor([1, 104, 64])]; tensor transpose_638_cast_fp16 = transpose(perm = transpose_638_perm_0, x = var_5913_cast_fp16_15)[name = string("transpose_3266")]; tensor reshape_957_cast_fp16 = reshape(shape = concat_3194, x = transpose_638_cast_fp16)[name = string("reshape_957_cast_fp16")]; tensor transpose_639_perm_0 = const()[name = string("transpose_639_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3195 = const()[name = string("concat_3195"), val = tensor([1, 64, 104])]; tensor transpose_639_cast_fp16 = transpose(perm = transpose_639_perm_0, x = var_5930_cast_fp16_15)[name = string("transpose_3265")]; tensor reshape_958_cast_fp16 = reshape(shape = concat_3195, x = transpose_639_cast_fp16)[name = string("reshape_958_cast_fp16")]; bool matmul_319_transpose_x_0 = const()[name = string("matmul_319_transpose_x_0"), val = bool(false)]; bool matmul_319_transpose_y_0 = const()[name = string("matmul_319_transpose_y_0"), val = bool(false)]; tensor matmul_319_cast_fp16 = matmul(transpose_x = matmul_319_transpose_x_0, transpose_y = matmul_319_transpose_y_0, x = reshape_957_cast_fp16, y = reshape_958_cast_fp16)[name = string("matmul_319_cast_fp16")]; tensor concat_3199 = const()[name = string("concat_3199"), val = tensor([1, 1, 104, 104])]; tensor reshape_959_cast_fp16 = reshape(shape = concat_3199, x = matmul_319_cast_fp16)[name = string("reshape_959_cast_fp16")]; tensor transpose_3007_perm_0 = const()[name = string("transpose_3007_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3007 = transpose(perm = transpose_3007_perm_0, x = reshape_959_cast_fp16)[name = string("transpose_3264")]; tensor w_1279_cast_fp16 = add(x = transpose_3007, y = transpose_2305)[name = string("w_1279_cast_fp16")]; tensor var_6089_cast_fp16 = softmax(axis = var_5857, x = w_1279_cast_fp16)[name = string("op_6089_cast_fp16")]; string var_6091_equation_0 = const()[name = string("op_6091_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6091_cast_fp16 = einsum(equation = var_6091_equation_0, values = (var_5947_cast_fp16_15, var_6089_cast_fp16))[name = string("op_6091_cast_fp16")]; bool input_163_interleave_0 = const()[name = string("input_163_interleave_0"), val = bool(false)]; tensor input_163_cast_fp16 = concat(axis = var_5857, interleave = input_163_interleave_0, values = (var_5971_cast_fp16, var_5979_cast_fp16, var_5987_cast_fp16, var_5995_cast_fp16, var_6003_cast_fp16, var_6011_cast_fp16, var_6019_cast_fp16, var_6027_cast_fp16, var_6035_cast_fp16, var_6043_cast_fp16, var_6051_cast_fp16, var_6059_cast_fp16, var_6067_cast_fp16, var_6075_cast_fp16, var_6083_cast_fp16, var_6091_cast_fp16))[name = string("input_163_cast_fp16")]; string var_6100_pad_type_0 = const()[name = string("op_6100_pad_type_0"), val = string("valid")]; tensor var_6100_strides_0 = const()[name = string("op_6100_strides_0"), val = tensor([1, 1])]; tensor var_6100_pad_0 = const()[name = string("op_6100_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6100_dilations_0 = const()[name = string("op_6100_dilations_0"), val = tensor([1, 1])]; int32 var_6100_groups_0 = const()[name = string("op_6100_groups_0"), val = int32(1)]; tensor layers_19_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_19_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(509051840)))]; tensor layers_19_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_19_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(511149056)))]; tensor var_6100_cast_fp16 = conv(bias = layers_19_self_attn_out_proj_bias_to_fp16, dilations = var_6100_dilations_0, groups = var_6100_groups_0, pad = var_6100_pad_0, pad_type = var_6100_pad_type_0, strides = var_6100_strides_0, weight = layers_19_self_attn_out_proj_weight_to_fp16, x = input_163_cast_fp16)[name = string("op_6100_cast_fp16")]; tensor x_207_cast_fp16 = add(x = x_203_cast_fp16, y = var_6100_cast_fp16)[name = string("x_207_cast_fp16")]; tensor mu_79_axes_0 = const()[name = string("mu_79_axes_0"), val = tensor([1])]; bool mu_79_keep_dims_0 = const()[name = string("mu_79_keep_dims_0"), val = bool(true)]; tensor mu_79_cast_fp16 = reduce_mean(axes = mu_79_axes_0, keep_dims = mu_79_keep_dims_0, x = x_207_cast_fp16)[name = string("mu_79_cast_fp16")]; tensor var_6106_cast_fp16 = sub(x = x_207_cast_fp16, y = mu_79_cast_fp16)[name = string("op_6106_cast_fp16")]; fp16 var_5860_promoted_1_to_fp16 = const()[name = string("op_5860_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_6107_cast_fp16 = pow(x = var_6106_cast_fp16, y = var_5860_promoted_1_to_fp16)[name = string("op_6107_cast_fp16")]; tensor var_79_axes_0 = const()[name = string("var_79_axes_0"), val = tensor([1])]; bool var_79_keep_dims_0 = const()[name = string("var_79_keep_dims_0"), val = bool(true)]; tensor var_79_cast_fp16 = reduce_mean(axes = var_79_axes_0, keep_dims = var_79_keep_dims_0, x = var_6107_cast_fp16)[name = string("var_79_cast_fp16")]; fp16 var_6111_to_fp16 = const()[name = string("op_6111_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_6112_cast_fp16 = add(x = var_79_cast_fp16, y = var_6111_to_fp16)[name = string("op_6112_cast_fp16")]; fp32 var_6113_epsilon_0 = const()[name = string("op_6113_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_6113_cast_fp16 = rsqrt(epsilon = var_6113_epsilon_0, x = var_6112_cast_fp16)[name = string("op_6113_cast_fp16")]; tensor x_209_cast_fp16 = mul(x = var_6106_cast_fp16, y = var_6113_cast_fp16)[name = string("x_209_cast_fp16")]; tensor input_165_gamma_0_to_fp16 = const()[name = string("input_165_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(511151168)))]; tensor input_165_beta_0_to_fp16 = const()[name = string("input_165_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(511153280)))]; fp16 input_165_epsilon_0_to_fp16 = const()[name = string("input_165_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_165_cast_fp16 = batch_norm(beta = input_165_beta_0_to_fp16, epsilon = input_165_epsilon_0_to_fp16, gamma = input_165_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_209_cast_fp16)[name = string("input_165_cast_fp16")]; string x_211_pad_type_0 = const()[name = string("x_211_pad_type_0"), val = string("valid")]; tensor x_211_strides_0 = const()[name = string("x_211_strides_0"), val = tensor([1, 1])]; tensor x_211_pad_0 = const()[name = string("x_211_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_211_dilations_0 = const()[name = string("x_211_dilations_0"), val = tensor([1, 1])]; int32 x_211_groups_0 = const()[name = string("x_211_groups_0"), val = int32(1)]; tensor layers_19_fc1_weight_to_fp16 = const()[name = string("layers_19_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(511155392)))]; tensor layers_19_fc1_bias_to_fp16 = const()[name = string("layers_19_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(519544064)))]; tensor x_211_cast_fp16 = conv(bias = layers_19_fc1_bias_to_fp16, dilations = x_211_dilations_0, groups = x_211_groups_0, pad = x_211_pad_0, pad_type = x_211_pad_type_0, strides = x_211_strides_0, weight = layers_19_fc1_weight_to_fp16, x = input_165_cast_fp16)[name = string("x_211_cast_fp16")]; fp16 var_6128_to_fp16 = const()[name = string("op_6128_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_6129_cast_fp16 = mul(x = x_211_cast_fp16, y = var_6128_to_fp16)[name = string("op_6129_cast_fp16")]; tensor var_6130_cast_fp16 = mul(x = var_6129_cast_fp16, y = x_211_cast_fp16)[name = string("op_6130_cast_fp16")]; tensor var_6131_cast_fp16 = mul(x = var_6130_cast_fp16, y = x_211_cast_fp16)[name = string("op_6131_cast_fp16")]; tensor var_6132_cast_fp16 = add(x = x_211_cast_fp16, y = var_6131_cast_fp16)[name = string("op_6132_cast_fp16")]; fp16 var_6133_to_fp16 = const()[name = string("op_6133_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_45_cast_fp16 = mul(x = var_6132_cast_fp16, y = var_6133_to_fp16)[name = string("u_45_cast_fp16")]; fp16 var_6135_to_fp16 = const()[name = string("op_6135_to_fp16"), val = fp16(0x1p-1)]; tensor var_6136_cast_fp16 = mul(x = x_211_cast_fp16, y = var_6135_to_fp16)[name = string("op_6136_cast_fp16")]; tensor var_6137_cast_fp16 = tanh(x = u_45_cast_fp16)[name = string("op_6137_cast_fp16")]; fp16 var_6138_to_fp16 = const()[name = string("op_6138_to_fp16"), val = fp16(0x1p+0)]; tensor var_6139_cast_fp16 = add(x = var_6137_cast_fp16, y = var_6138_to_fp16)[name = string("op_6139_cast_fp16")]; tensor input_167_cast_fp16 = mul(x = var_6136_cast_fp16, y = var_6139_cast_fp16)[name = string("input_167_cast_fp16")]; string h_39_pad_type_0 = const()[name = string("h_39_pad_type_0"), val = string("valid")]; tensor h_39_strides_0 = const()[name = string("h_39_strides_0"), val = tensor([1, 1])]; tensor h_39_pad_0 = const()[name = string("h_39_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_39_dilations_0 = const()[name = string("h_39_dilations_0"), val = tensor([1, 1])]; int32 h_39_groups_0 = const()[name = string("h_39_groups_0"), val = int32(1)]; tensor layers_19_fc2_weight_to_fp16 = const()[name = string("layers_19_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(519552320)))]; tensor layers_19_fc2_bias_to_fp16 = const()[name = string("layers_19_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(527940992)))]; tensor h_39_cast_fp16 = conv(bias = layers_19_fc2_bias_to_fp16, dilations = h_39_dilations_0, groups = h_39_groups_0, pad = h_39_pad_0, pad_type = h_39_pad_type_0, strides = h_39_strides_0, weight = layers_19_fc2_weight_to_fp16, x = input_167_cast_fp16)[name = string("h_39_cast_fp16")]; tensor x_213_cast_fp16 = add(x = x_207_cast_fp16, y = h_39_cast_fp16)[name = string("x_213_cast_fp16")]; int32 var_6155 = const()[name = string("op_6155"), val = int32(1)]; tensor mu_81_axes_0 = const()[name = string("mu_81_axes_0"), val = tensor([1])]; bool mu_81_keep_dims_0 = const()[name = string("mu_81_keep_dims_0"), val = bool(true)]; tensor mu_81_cast_fp16 = reduce_mean(axes = mu_81_axes_0, keep_dims = mu_81_keep_dims_0, x = x_213_cast_fp16)[name = string("mu_81_cast_fp16")]; tensor var_6169_cast_fp16 = sub(x = x_213_cast_fp16, y = mu_81_cast_fp16)[name = string("op_6169_cast_fp16")]; fp16 var_6158_promoted_to_fp16 = const()[name = string("op_6158_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_6170_cast_fp16 = pow(x = var_6169_cast_fp16, y = var_6158_promoted_to_fp16)[name = string("op_6170_cast_fp16")]; tensor var_81_axes_0 = const()[name = string("var_81_axes_0"), val = tensor([1])]; bool var_81_keep_dims_0 = const()[name = string("var_81_keep_dims_0"), val = bool(true)]; tensor var_81_cast_fp16 = reduce_mean(axes = var_81_axes_0, keep_dims = var_81_keep_dims_0, x = var_6170_cast_fp16)[name = string("var_81_cast_fp16")]; fp16 var_6174_to_fp16 = const()[name = string("op_6174_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_6175_cast_fp16 = add(x = var_81_cast_fp16, y = var_6174_to_fp16)[name = string("op_6175_cast_fp16")]; fp32 var_6176_epsilon_0 = const()[name = string("op_6176_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_6176_cast_fp16 = rsqrt(epsilon = var_6176_epsilon_0, x = var_6175_cast_fp16)[name = string("op_6176_cast_fp16")]; tensor x_215_cast_fp16 = mul(x = var_6169_cast_fp16, y = var_6176_cast_fp16)[name = string("x_215_cast_fp16")]; tensor input_169_gamma_0_to_fp16 = const()[name = string("input_169_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(527943104)))]; tensor input_169_beta_0_to_fp16 = const()[name = string("input_169_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(527945216)))]; fp16 input_169_epsilon_0_to_fp16 = const()[name = string("input_169_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_169_cast_fp16 = batch_norm(beta = input_169_beta_0_to_fp16, epsilon = input_169_epsilon_0_to_fp16, gamma = input_169_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_215_cast_fp16)[name = string("input_169_cast_fp16")]; string var_6194_pad_type_0 = const()[name = string("op_6194_pad_type_0"), val = string("valid")]; tensor var_6194_strides_0 = const()[name = string("op_6194_strides_0"), val = tensor([1, 1])]; tensor var_6194_pad_0 = const()[name = string("op_6194_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6194_dilations_0 = const()[name = string("op_6194_dilations_0"), val = tensor([1, 1])]; int32 var_6194_groups_0 = const()[name = string("op_6194_groups_0"), val = int32(1)]; tensor var_6196_weight_0_to_fp16 = const()[name = string("op_6196_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(527947328)))]; tensor var_6196_bias_0_to_fp16 = const()[name = string("op_6196_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(530044544)))]; tensor var_6196_cast_fp16 = conv(bias = var_6196_bias_0_to_fp16, dilations = var_6194_dilations_0, groups = var_6194_groups_0, pad = var_6194_pad_0, pad_type = var_6194_pad_type_0, strides = var_6194_strides_0, weight = var_6196_weight_0_to_fp16, x = input_169_cast_fp16)[name = string("op_6196_cast_fp16")]; string var_6203_pad_type_0 = const()[name = string("op_6203_pad_type_0"), val = string("valid")]; tensor var_6203_strides_0 = const()[name = string("op_6203_strides_0"), val = tensor([1, 1])]; tensor var_6203_pad_0 = const()[name = string("op_6203_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6203_dilations_0 = const()[name = string("op_6203_dilations_0"), val = tensor([1, 1])]; int32 var_6203_groups_0 = const()[name = string("op_6203_groups_0"), val = int32(1)]; tensor layers_20_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_20_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(530046656)))]; tensor layers_20_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_20_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(532143872)))]; tensor var_6203_cast_fp16 = conv(bias = layers_20_self_attn_k_proj_bias_to_fp16, dilations = var_6203_dilations_0, groups = var_6203_groups_0, pad = var_6203_pad_0, pad_type = var_6203_pad_type_0, strides = var_6203_strides_0, weight = layers_20_self_attn_k_proj_weight_to_fp16, x = input_169_cast_fp16)[name = string("op_6203_cast_fp16")]; string var_6210_pad_type_0 = const()[name = string("op_6210_pad_type_0"), val = string("valid")]; tensor var_6210_strides_0 = const()[name = string("op_6210_strides_0"), val = tensor([1, 1])]; tensor var_6210_pad_0 = const()[name = string("op_6210_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6210_dilations_0 = const()[name = string("op_6210_dilations_0"), val = tensor([1, 1])]; int32 var_6210_groups_0 = const()[name = string("op_6210_groups_0"), val = int32(1)]; tensor layers_20_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_20_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(532145984)))]; tensor layers_20_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_20_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(534243200)))]; tensor var_6210_cast_fp16 = conv(bias = layers_20_self_attn_v_proj_bias_to_fp16, dilations = var_6210_dilations_0, groups = var_6210_groups_0, pad = var_6210_pad_0, pad_type = var_6210_pad_type_0, strides = var_6210_strides_0, weight = layers_20_self_attn_v_proj_weight_to_fp16, x = input_169_cast_fp16)[name = string("op_6210_cast_fp16")]; tensor tile_60 = const()[name = string("tile_60"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(534245312)))]; int32 var_6211_axis_0 = const()[name = string("op_6211_axis_0"), val = int32(1)]; tensor var_6211_cast_fp16_0, tensor var_6211_cast_fp16_1, tensor var_6211_cast_fp16_2, tensor var_6211_cast_fp16_3, tensor var_6211_cast_fp16_4, tensor var_6211_cast_fp16_5, tensor var_6211_cast_fp16_6, tensor var_6211_cast_fp16_7, tensor var_6211_cast_fp16_8, tensor var_6211_cast_fp16_9, tensor var_6211_cast_fp16_10, tensor var_6211_cast_fp16_11, tensor var_6211_cast_fp16_12, tensor var_6211_cast_fp16_13, tensor var_6211_cast_fp16_14, tensor var_6211_cast_fp16_15 = split(axis = var_6211_axis_0, split_sizes = tile_60, x = var_6196_cast_fp16)[name = string("op_6211_cast_fp16")]; tensor tile_61 = const()[name = string("tile_61"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(534245440)))]; int32 var_6228_axis_0 = const()[name = string("op_6228_axis_0"), val = int32(1)]; tensor var_6228_cast_fp16_0, tensor var_6228_cast_fp16_1, tensor var_6228_cast_fp16_2, tensor var_6228_cast_fp16_3, tensor var_6228_cast_fp16_4, tensor var_6228_cast_fp16_5, tensor var_6228_cast_fp16_6, tensor var_6228_cast_fp16_7, tensor var_6228_cast_fp16_8, tensor var_6228_cast_fp16_9, tensor var_6228_cast_fp16_10, tensor var_6228_cast_fp16_11, tensor var_6228_cast_fp16_12, tensor var_6228_cast_fp16_13, tensor var_6228_cast_fp16_14, tensor var_6228_cast_fp16_15 = split(axis = var_6228_axis_0, split_sizes = tile_61, x = var_6203_cast_fp16)[name = string("op_6228_cast_fp16")]; tensor tile_62 = const()[name = string("tile_62"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(534245568)))]; int32 var_6245_axis_0 = const()[name = string("op_6245_axis_0"), val = int32(1)]; tensor var_6245_cast_fp16_0, tensor var_6245_cast_fp16_1, tensor var_6245_cast_fp16_2, tensor var_6245_cast_fp16_3, tensor var_6245_cast_fp16_4, tensor var_6245_cast_fp16_5, tensor var_6245_cast_fp16_6, tensor var_6245_cast_fp16_7, tensor var_6245_cast_fp16_8, tensor var_6245_cast_fp16_9, tensor var_6245_cast_fp16_10, tensor var_6245_cast_fp16_11, tensor var_6245_cast_fp16_12, tensor var_6245_cast_fp16_13, tensor var_6245_cast_fp16_14, tensor var_6245_cast_fp16_15 = split(axis = var_6245_axis_0, split_sizes = tile_62, x = var_6210_cast_fp16)[name = string("op_6245_cast_fp16")]; tensor transpose_640_perm_0 = const()[name = string("transpose_640_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3204 = const()[name = string("concat_3204"), val = tensor([1, 104, 64])]; tensor transpose_640_cast_fp16 = transpose(perm = transpose_640_perm_0, x = var_6211_cast_fp16_0)[name = string("transpose_3263")]; tensor reshape_960_cast_fp16 = reshape(shape = concat_3204, x = transpose_640_cast_fp16)[name = string("reshape_960_cast_fp16")]; tensor transpose_641_perm_0 = const()[name = string("transpose_641_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3205 = const()[name = string("concat_3205"), val = tensor([1, 64, 104])]; tensor transpose_641_cast_fp16 = transpose(perm = transpose_641_perm_0, x = var_6228_cast_fp16_0)[name = string("transpose_3262")]; tensor reshape_961_cast_fp16 = reshape(shape = concat_3205, x = transpose_641_cast_fp16)[name = string("reshape_961_cast_fp16")]; bool matmul_320_transpose_x_0 = const()[name = string("matmul_320_transpose_x_0"), val = bool(false)]; bool matmul_320_transpose_y_0 = const()[name = string("matmul_320_transpose_y_0"), val = bool(false)]; tensor matmul_320_cast_fp16 = matmul(transpose_x = matmul_320_transpose_x_0, transpose_y = matmul_320_transpose_y_0, x = reshape_960_cast_fp16, y = reshape_961_cast_fp16)[name = string("matmul_320_cast_fp16")]; tensor concat_3209 = const()[name = string("concat_3209"), val = tensor([1, 1, 104, 104])]; tensor reshape_962_cast_fp16 = reshape(shape = concat_3209, x = matmul_320_cast_fp16)[name = string("reshape_962_cast_fp16")]; tensor transpose_3008_perm_0 = const()[name = string("transpose_3008_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3008 = transpose(perm = transpose_3008_perm_0, x = reshape_962_cast_fp16)[name = string("transpose_3261")]; tensor w_1283_cast_fp16 = add(x = transpose_3008, y = transpose_2305)[name = string("w_1283_cast_fp16")]; tensor var_6267_cast_fp16 = softmax(axis = var_6155, x = w_1283_cast_fp16)[name = string("op_6267_cast_fp16")]; string var_6269_equation_0 = const()[name = string("op_6269_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6269_cast_fp16 = einsum(equation = var_6269_equation_0, values = (var_6245_cast_fp16_0, var_6267_cast_fp16))[name = string("op_6269_cast_fp16")]; tensor transpose_642_perm_0 = const()[name = string("transpose_642_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3214 = const()[name = string("concat_3214"), val = tensor([1, 104, 64])]; tensor transpose_642_cast_fp16 = transpose(perm = transpose_642_perm_0, x = var_6211_cast_fp16_1)[name = string("transpose_3260")]; tensor reshape_963_cast_fp16 = reshape(shape = concat_3214, x = transpose_642_cast_fp16)[name = string("reshape_963_cast_fp16")]; tensor transpose_643_perm_0 = const()[name = string("transpose_643_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3215 = const()[name = string("concat_3215"), val = tensor([1, 64, 104])]; tensor transpose_643_cast_fp16 = transpose(perm = transpose_643_perm_0, x = var_6228_cast_fp16_1)[name = string("transpose_3259")]; tensor reshape_964_cast_fp16 = reshape(shape = concat_3215, x = transpose_643_cast_fp16)[name = string("reshape_964_cast_fp16")]; bool matmul_321_transpose_x_0 = const()[name = string("matmul_321_transpose_x_0"), val = bool(false)]; bool matmul_321_transpose_y_0 = const()[name = string("matmul_321_transpose_y_0"), val = bool(false)]; tensor matmul_321_cast_fp16 = matmul(transpose_x = matmul_321_transpose_x_0, transpose_y = matmul_321_transpose_y_0, x = reshape_963_cast_fp16, y = reshape_964_cast_fp16)[name = string("matmul_321_cast_fp16")]; tensor concat_3219 = const()[name = string("concat_3219"), val = tensor([1, 1, 104, 104])]; tensor reshape_965_cast_fp16 = reshape(shape = concat_3219, x = matmul_321_cast_fp16)[name = string("reshape_965_cast_fp16")]; tensor transpose_3009_perm_0 = const()[name = string("transpose_3009_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3009 = transpose(perm = transpose_3009_perm_0, x = reshape_965_cast_fp16)[name = string("transpose_3258")]; tensor w_1287_cast_fp16 = add(x = transpose_3009, y = transpose_2305)[name = string("w_1287_cast_fp16")]; tensor var_6275_cast_fp16 = softmax(axis = var_6155, x = w_1287_cast_fp16)[name = string("op_6275_cast_fp16")]; string var_6277_equation_0 = const()[name = string("op_6277_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6277_cast_fp16 = einsum(equation = var_6277_equation_0, values = (var_6245_cast_fp16_1, var_6275_cast_fp16))[name = string("op_6277_cast_fp16")]; tensor transpose_644_perm_0 = const()[name = string("transpose_644_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3224 = const()[name = string("concat_3224"), val = tensor([1, 104, 64])]; tensor transpose_644_cast_fp16 = transpose(perm = transpose_644_perm_0, x = var_6211_cast_fp16_2)[name = string("transpose_3257")]; tensor reshape_966_cast_fp16 = reshape(shape = concat_3224, x = transpose_644_cast_fp16)[name = string("reshape_966_cast_fp16")]; tensor transpose_645_perm_0 = const()[name = string("transpose_645_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3225 = const()[name = string("concat_3225"), val = tensor([1, 64, 104])]; tensor transpose_645_cast_fp16 = transpose(perm = transpose_645_perm_0, x = var_6228_cast_fp16_2)[name = string("transpose_3256")]; tensor reshape_967_cast_fp16 = reshape(shape = concat_3225, x = transpose_645_cast_fp16)[name = string("reshape_967_cast_fp16")]; bool matmul_322_transpose_x_0 = const()[name = string("matmul_322_transpose_x_0"), val = bool(false)]; bool matmul_322_transpose_y_0 = const()[name = string("matmul_322_transpose_y_0"), val = bool(false)]; tensor matmul_322_cast_fp16 = matmul(transpose_x = matmul_322_transpose_x_0, transpose_y = matmul_322_transpose_y_0, x = reshape_966_cast_fp16, y = reshape_967_cast_fp16)[name = string("matmul_322_cast_fp16")]; tensor concat_3229 = const()[name = string("concat_3229"), val = tensor([1, 1, 104, 104])]; tensor reshape_968_cast_fp16 = reshape(shape = concat_3229, x = matmul_322_cast_fp16)[name = string("reshape_968_cast_fp16")]; tensor transpose_3010_perm_0 = const()[name = string("transpose_3010_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3010 = transpose(perm = transpose_3010_perm_0, x = reshape_968_cast_fp16)[name = string("transpose_3255")]; tensor w_1291_cast_fp16 = add(x = transpose_3010, y = transpose_2305)[name = string("w_1291_cast_fp16")]; tensor var_6283_cast_fp16 = softmax(axis = var_6155, x = w_1291_cast_fp16)[name = string("op_6283_cast_fp16")]; string var_6285_equation_0 = const()[name = string("op_6285_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6285_cast_fp16 = einsum(equation = var_6285_equation_0, values = (var_6245_cast_fp16_2, var_6283_cast_fp16))[name = string("op_6285_cast_fp16")]; tensor transpose_646_perm_0 = const()[name = string("transpose_646_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3234 = const()[name = string("concat_3234"), val = tensor([1, 104, 64])]; tensor transpose_646_cast_fp16 = transpose(perm = transpose_646_perm_0, x = var_6211_cast_fp16_3)[name = string("transpose_3254")]; tensor reshape_969_cast_fp16 = reshape(shape = concat_3234, x = transpose_646_cast_fp16)[name = string("reshape_969_cast_fp16")]; tensor transpose_647_perm_0 = const()[name = string("transpose_647_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3235 = const()[name = string("concat_3235"), val = tensor([1, 64, 104])]; tensor transpose_647_cast_fp16 = transpose(perm = transpose_647_perm_0, x = var_6228_cast_fp16_3)[name = string("transpose_3253")]; tensor reshape_970_cast_fp16 = reshape(shape = concat_3235, x = transpose_647_cast_fp16)[name = string("reshape_970_cast_fp16")]; bool matmul_323_transpose_x_0 = const()[name = string("matmul_323_transpose_x_0"), val = bool(false)]; bool matmul_323_transpose_y_0 = const()[name = string("matmul_323_transpose_y_0"), val = bool(false)]; tensor matmul_323_cast_fp16 = matmul(transpose_x = matmul_323_transpose_x_0, transpose_y = matmul_323_transpose_y_0, x = reshape_969_cast_fp16, y = reshape_970_cast_fp16)[name = string("matmul_323_cast_fp16")]; tensor concat_3239 = const()[name = string("concat_3239"), val = tensor([1, 1, 104, 104])]; tensor reshape_971_cast_fp16 = reshape(shape = concat_3239, x = matmul_323_cast_fp16)[name = string("reshape_971_cast_fp16")]; tensor transpose_3011_perm_0 = const()[name = string("transpose_3011_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3011 = transpose(perm = transpose_3011_perm_0, x = reshape_971_cast_fp16)[name = string("transpose_3252")]; tensor w_1295_cast_fp16 = add(x = transpose_3011, y = transpose_2305)[name = string("w_1295_cast_fp16")]; tensor var_6291_cast_fp16 = softmax(axis = var_6155, x = w_1295_cast_fp16)[name = string("op_6291_cast_fp16")]; string var_6293_equation_0 = const()[name = string("op_6293_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6293_cast_fp16 = einsum(equation = var_6293_equation_0, values = (var_6245_cast_fp16_3, var_6291_cast_fp16))[name = string("op_6293_cast_fp16")]; tensor transpose_648_perm_0 = const()[name = string("transpose_648_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3244 = const()[name = string("concat_3244"), val = tensor([1, 104, 64])]; tensor transpose_648_cast_fp16 = transpose(perm = transpose_648_perm_0, x = var_6211_cast_fp16_4)[name = string("transpose_3251")]; tensor reshape_972_cast_fp16 = reshape(shape = concat_3244, x = transpose_648_cast_fp16)[name = string("reshape_972_cast_fp16")]; tensor transpose_649_perm_0 = const()[name = string("transpose_649_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3245 = const()[name = string("concat_3245"), val = tensor([1, 64, 104])]; tensor transpose_649_cast_fp16 = transpose(perm = transpose_649_perm_0, x = var_6228_cast_fp16_4)[name = string("transpose_3250")]; tensor reshape_973_cast_fp16 = reshape(shape = concat_3245, x = transpose_649_cast_fp16)[name = string("reshape_973_cast_fp16")]; bool matmul_324_transpose_x_0 = const()[name = string("matmul_324_transpose_x_0"), val = bool(false)]; bool matmul_324_transpose_y_0 = const()[name = string("matmul_324_transpose_y_0"), val = bool(false)]; tensor matmul_324_cast_fp16 = matmul(transpose_x = matmul_324_transpose_x_0, transpose_y = matmul_324_transpose_y_0, x = reshape_972_cast_fp16, y = reshape_973_cast_fp16)[name = string("matmul_324_cast_fp16")]; tensor concat_3249 = const()[name = string("concat_3249"), val = tensor([1, 1, 104, 104])]; tensor reshape_974_cast_fp16 = reshape(shape = concat_3249, x = matmul_324_cast_fp16)[name = string("reshape_974_cast_fp16")]; tensor transpose_3012_perm_0 = const()[name = string("transpose_3012_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3012 = transpose(perm = transpose_3012_perm_0, x = reshape_974_cast_fp16)[name = string("transpose_3249")]; tensor w_1299_cast_fp16 = add(x = transpose_3012, y = transpose_2305)[name = string("w_1299_cast_fp16")]; tensor var_6299_cast_fp16 = softmax(axis = var_6155, x = w_1299_cast_fp16)[name = string("op_6299_cast_fp16")]; string var_6301_equation_0 = const()[name = string("op_6301_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6301_cast_fp16 = einsum(equation = var_6301_equation_0, values = (var_6245_cast_fp16_4, var_6299_cast_fp16))[name = string("op_6301_cast_fp16")]; tensor transpose_650_perm_0 = const()[name = string("transpose_650_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3254 = const()[name = string("concat_3254"), val = tensor([1, 104, 64])]; tensor transpose_650_cast_fp16 = transpose(perm = transpose_650_perm_0, x = var_6211_cast_fp16_5)[name = string("transpose_3248")]; tensor reshape_975_cast_fp16 = reshape(shape = concat_3254, x = transpose_650_cast_fp16)[name = string("reshape_975_cast_fp16")]; tensor transpose_651_perm_0 = const()[name = string("transpose_651_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3255 = const()[name = string("concat_3255"), val = tensor([1, 64, 104])]; tensor transpose_651_cast_fp16 = transpose(perm = transpose_651_perm_0, x = var_6228_cast_fp16_5)[name = string("transpose_3247")]; tensor reshape_976_cast_fp16 = reshape(shape = concat_3255, x = transpose_651_cast_fp16)[name = string("reshape_976_cast_fp16")]; bool matmul_325_transpose_x_0 = const()[name = string("matmul_325_transpose_x_0"), val = bool(false)]; bool matmul_325_transpose_y_0 = const()[name = string("matmul_325_transpose_y_0"), val = bool(false)]; tensor matmul_325_cast_fp16 = matmul(transpose_x = matmul_325_transpose_x_0, transpose_y = matmul_325_transpose_y_0, x = reshape_975_cast_fp16, y = reshape_976_cast_fp16)[name = string("matmul_325_cast_fp16")]; tensor concat_3259 = const()[name = string("concat_3259"), val = tensor([1, 1, 104, 104])]; tensor reshape_977_cast_fp16 = reshape(shape = concat_3259, x = matmul_325_cast_fp16)[name = string("reshape_977_cast_fp16")]; tensor transpose_3013_perm_0 = const()[name = string("transpose_3013_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3013 = transpose(perm = transpose_3013_perm_0, x = reshape_977_cast_fp16)[name = string("transpose_3246")]; tensor w_1303_cast_fp16 = add(x = transpose_3013, y = transpose_2305)[name = string("w_1303_cast_fp16")]; tensor var_6307_cast_fp16 = softmax(axis = var_6155, x = w_1303_cast_fp16)[name = string("op_6307_cast_fp16")]; string var_6309_equation_0 = const()[name = string("op_6309_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6309_cast_fp16 = einsum(equation = var_6309_equation_0, values = (var_6245_cast_fp16_5, var_6307_cast_fp16))[name = string("op_6309_cast_fp16")]; tensor transpose_652_perm_0 = const()[name = string("transpose_652_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3264 = const()[name = string("concat_3264"), val = tensor([1, 104, 64])]; tensor transpose_652_cast_fp16 = transpose(perm = transpose_652_perm_0, x = var_6211_cast_fp16_6)[name = string("transpose_3245")]; tensor reshape_978_cast_fp16 = reshape(shape = concat_3264, x = transpose_652_cast_fp16)[name = string("reshape_978_cast_fp16")]; tensor transpose_653_perm_0 = const()[name = string("transpose_653_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3265 = const()[name = string("concat_3265"), val = tensor([1, 64, 104])]; tensor transpose_653_cast_fp16 = transpose(perm = transpose_653_perm_0, x = var_6228_cast_fp16_6)[name = string("transpose_3244")]; tensor reshape_979_cast_fp16 = reshape(shape = concat_3265, x = transpose_653_cast_fp16)[name = string("reshape_979_cast_fp16")]; bool matmul_326_transpose_x_0 = const()[name = string("matmul_326_transpose_x_0"), val = bool(false)]; bool matmul_326_transpose_y_0 = const()[name = string("matmul_326_transpose_y_0"), val = bool(false)]; tensor matmul_326_cast_fp16 = matmul(transpose_x = matmul_326_transpose_x_0, transpose_y = matmul_326_transpose_y_0, x = reshape_978_cast_fp16, y = reshape_979_cast_fp16)[name = string("matmul_326_cast_fp16")]; tensor concat_3269 = const()[name = string("concat_3269"), val = tensor([1, 1, 104, 104])]; tensor reshape_980_cast_fp16 = reshape(shape = concat_3269, x = matmul_326_cast_fp16)[name = string("reshape_980_cast_fp16")]; tensor transpose_3014_perm_0 = const()[name = string("transpose_3014_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3014 = transpose(perm = transpose_3014_perm_0, x = reshape_980_cast_fp16)[name = string("transpose_3243")]; tensor w_1307_cast_fp16 = add(x = transpose_3014, y = transpose_2305)[name = string("w_1307_cast_fp16")]; tensor var_6315_cast_fp16 = softmax(axis = var_6155, x = w_1307_cast_fp16)[name = string("op_6315_cast_fp16")]; string var_6317_equation_0 = const()[name = string("op_6317_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6317_cast_fp16 = einsum(equation = var_6317_equation_0, values = (var_6245_cast_fp16_6, var_6315_cast_fp16))[name = string("op_6317_cast_fp16")]; tensor transpose_654_perm_0 = const()[name = string("transpose_654_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3274 = const()[name = string("concat_3274"), val = tensor([1, 104, 64])]; tensor transpose_654_cast_fp16 = transpose(perm = transpose_654_perm_0, x = var_6211_cast_fp16_7)[name = string("transpose_3242")]; tensor reshape_981_cast_fp16 = reshape(shape = concat_3274, x = transpose_654_cast_fp16)[name = string("reshape_981_cast_fp16")]; tensor transpose_655_perm_0 = const()[name = string("transpose_655_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3275 = const()[name = string("concat_3275"), val = tensor([1, 64, 104])]; tensor transpose_655_cast_fp16 = transpose(perm = transpose_655_perm_0, x = var_6228_cast_fp16_7)[name = string("transpose_3241")]; tensor reshape_982_cast_fp16 = reshape(shape = concat_3275, x = transpose_655_cast_fp16)[name = string("reshape_982_cast_fp16")]; bool matmul_327_transpose_x_0 = const()[name = string("matmul_327_transpose_x_0"), val = bool(false)]; bool matmul_327_transpose_y_0 = const()[name = string("matmul_327_transpose_y_0"), val = bool(false)]; tensor matmul_327_cast_fp16 = matmul(transpose_x = matmul_327_transpose_x_0, transpose_y = matmul_327_transpose_y_0, x = reshape_981_cast_fp16, y = reshape_982_cast_fp16)[name = string("matmul_327_cast_fp16")]; tensor concat_3279 = const()[name = string("concat_3279"), val = tensor([1, 1, 104, 104])]; tensor reshape_983_cast_fp16 = reshape(shape = concat_3279, x = matmul_327_cast_fp16)[name = string("reshape_983_cast_fp16")]; tensor transpose_3015_perm_0 = const()[name = string("transpose_3015_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3015 = transpose(perm = transpose_3015_perm_0, x = reshape_983_cast_fp16)[name = string("transpose_3240")]; tensor w_1311_cast_fp16 = add(x = transpose_3015, y = transpose_2305)[name = string("w_1311_cast_fp16")]; tensor var_6323_cast_fp16 = softmax(axis = var_6155, x = w_1311_cast_fp16)[name = string("op_6323_cast_fp16")]; string var_6325_equation_0 = const()[name = string("op_6325_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6325_cast_fp16 = einsum(equation = var_6325_equation_0, values = (var_6245_cast_fp16_7, var_6323_cast_fp16))[name = string("op_6325_cast_fp16")]; tensor transpose_656_perm_0 = const()[name = string("transpose_656_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3284 = const()[name = string("concat_3284"), val = tensor([1, 104, 64])]; tensor transpose_656_cast_fp16 = transpose(perm = transpose_656_perm_0, x = var_6211_cast_fp16_8)[name = string("transpose_3239")]; tensor reshape_984_cast_fp16 = reshape(shape = concat_3284, x = transpose_656_cast_fp16)[name = string("reshape_984_cast_fp16")]; tensor transpose_657_perm_0 = const()[name = string("transpose_657_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3285 = const()[name = string("concat_3285"), val = tensor([1, 64, 104])]; tensor transpose_657_cast_fp16 = transpose(perm = transpose_657_perm_0, x = var_6228_cast_fp16_8)[name = string("transpose_3238")]; tensor reshape_985_cast_fp16 = reshape(shape = concat_3285, x = transpose_657_cast_fp16)[name = string("reshape_985_cast_fp16")]; bool matmul_328_transpose_x_0 = const()[name = string("matmul_328_transpose_x_0"), val = bool(false)]; bool matmul_328_transpose_y_0 = const()[name = string("matmul_328_transpose_y_0"), val = bool(false)]; tensor matmul_328_cast_fp16 = matmul(transpose_x = matmul_328_transpose_x_0, transpose_y = matmul_328_transpose_y_0, x = reshape_984_cast_fp16, y = reshape_985_cast_fp16)[name = string("matmul_328_cast_fp16")]; tensor concat_3289 = const()[name = string("concat_3289"), val = tensor([1, 1, 104, 104])]; tensor reshape_986_cast_fp16 = reshape(shape = concat_3289, x = matmul_328_cast_fp16)[name = string("reshape_986_cast_fp16")]; tensor transpose_3016_perm_0 = const()[name = string("transpose_3016_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3016 = transpose(perm = transpose_3016_perm_0, x = reshape_986_cast_fp16)[name = string("transpose_3237")]; tensor w_1315_cast_fp16 = add(x = transpose_3016, y = transpose_2305)[name = string("w_1315_cast_fp16")]; tensor var_6331_cast_fp16 = softmax(axis = var_6155, x = w_1315_cast_fp16)[name = string("op_6331_cast_fp16")]; string var_6333_equation_0 = const()[name = string("op_6333_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6333_cast_fp16 = einsum(equation = var_6333_equation_0, values = (var_6245_cast_fp16_8, var_6331_cast_fp16))[name = string("op_6333_cast_fp16")]; tensor transpose_658_perm_0 = const()[name = string("transpose_658_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3294 = const()[name = string("concat_3294"), val = tensor([1, 104, 64])]; tensor transpose_658_cast_fp16 = transpose(perm = transpose_658_perm_0, x = var_6211_cast_fp16_9)[name = string("transpose_3236")]; tensor reshape_987_cast_fp16 = reshape(shape = concat_3294, x = transpose_658_cast_fp16)[name = string("reshape_987_cast_fp16")]; tensor transpose_659_perm_0 = const()[name = string("transpose_659_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3295 = const()[name = string("concat_3295"), val = tensor([1, 64, 104])]; tensor transpose_659_cast_fp16 = transpose(perm = transpose_659_perm_0, x = var_6228_cast_fp16_9)[name = string("transpose_3235")]; tensor reshape_988_cast_fp16 = reshape(shape = concat_3295, x = transpose_659_cast_fp16)[name = string("reshape_988_cast_fp16")]; bool matmul_329_transpose_x_0 = const()[name = string("matmul_329_transpose_x_0"), val = bool(false)]; bool matmul_329_transpose_y_0 = const()[name = string("matmul_329_transpose_y_0"), val = bool(false)]; tensor matmul_329_cast_fp16 = matmul(transpose_x = matmul_329_transpose_x_0, transpose_y = matmul_329_transpose_y_0, x = reshape_987_cast_fp16, y = reshape_988_cast_fp16)[name = string("matmul_329_cast_fp16")]; tensor concat_3299 = const()[name = string("concat_3299"), val = tensor([1, 1, 104, 104])]; tensor reshape_989_cast_fp16 = reshape(shape = concat_3299, x = matmul_329_cast_fp16)[name = string("reshape_989_cast_fp16")]; tensor transpose_3017_perm_0 = const()[name = string("transpose_3017_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3017 = transpose(perm = transpose_3017_perm_0, x = reshape_989_cast_fp16)[name = string("transpose_3234")]; tensor w_1319_cast_fp16 = add(x = transpose_3017, y = transpose_2305)[name = string("w_1319_cast_fp16")]; tensor var_6339_cast_fp16 = softmax(axis = var_6155, x = w_1319_cast_fp16)[name = string("op_6339_cast_fp16")]; string var_6341_equation_0 = const()[name = string("op_6341_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6341_cast_fp16 = einsum(equation = var_6341_equation_0, values = (var_6245_cast_fp16_9, var_6339_cast_fp16))[name = string("op_6341_cast_fp16")]; tensor transpose_660_perm_0 = const()[name = string("transpose_660_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3304 = const()[name = string("concat_3304"), val = tensor([1, 104, 64])]; tensor transpose_660_cast_fp16 = transpose(perm = transpose_660_perm_0, x = var_6211_cast_fp16_10)[name = string("transpose_3233")]; tensor reshape_990_cast_fp16 = reshape(shape = concat_3304, x = transpose_660_cast_fp16)[name = string("reshape_990_cast_fp16")]; tensor transpose_661_perm_0 = const()[name = string("transpose_661_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3305 = const()[name = string("concat_3305"), val = tensor([1, 64, 104])]; tensor transpose_661_cast_fp16 = transpose(perm = transpose_661_perm_0, x = var_6228_cast_fp16_10)[name = string("transpose_3232")]; tensor reshape_991_cast_fp16 = reshape(shape = concat_3305, x = transpose_661_cast_fp16)[name = string("reshape_991_cast_fp16")]; bool matmul_330_transpose_x_0 = const()[name = string("matmul_330_transpose_x_0"), val = bool(false)]; bool matmul_330_transpose_y_0 = const()[name = string("matmul_330_transpose_y_0"), val = bool(false)]; tensor matmul_330_cast_fp16 = matmul(transpose_x = matmul_330_transpose_x_0, transpose_y = matmul_330_transpose_y_0, x = reshape_990_cast_fp16, y = reshape_991_cast_fp16)[name = string("matmul_330_cast_fp16")]; tensor concat_3309 = const()[name = string("concat_3309"), val = tensor([1, 1, 104, 104])]; tensor reshape_992_cast_fp16 = reshape(shape = concat_3309, x = matmul_330_cast_fp16)[name = string("reshape_992_cast_fp16")]; tensor transpose_3018_perm_0 = const()[name = string("transpose_3018_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3018 = transpose(perm = transpose_3018_perm_0, x = reshape_992_cast_fp16)[name = string("transpose_3231")]; tensor w_1323_cast_fp16 = add(x = transpose_3018, y = transpose_2305)[name = string("w_1323_cast_fp16")]; tensor var_6347_cast_fp16 = softmax(axis = var_6155, x = w_1323_cast_fp16)[name = string("op_6347_cast_fp16")]; string var_6349_equation_0 = const()[name = string("op_6349_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6349_cast_fp16 = einsum(equation = var_6349_equation_0, values = (var_6245_cast_fp16_10, var_6347_cast_fp16))[name = string("op_6349_cast_fp16")]; tensor transpose_662_perm_0 = const()[name = string("transpose_662_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3314 = const()[name = string("concat_3314"), val = tensor([1, 104, 64])]; tensor transpose_662_cast_fp16 = transpose(perm = transpose_662_perm_0, x = var_6211_cast_fp16_11)[name = string("transpose_3230")]; tensor reshape_993_cast_fp16 = reshape(shape = concat_3314, x = transpose_662_cast_fp16)[name = string("reshape_993_cast_fp16")]; tensor transpose_663_perm_0 = const()[name = string("transpose_663_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3315 = const()[name = string("concat_3315"), val = tensor([1, 64, 104])]; tensor transpose_663_cast_fp16 = transpose(perm = transpose_663_perm_0, x = var_6228_cast_fp16_11)[name = string("transpose_3229")]; tensor reshape_994_cast_fp16 = reshape(shape = concat_3315, x = transpose_663_cast_fp16)[name = string("reshape_994_cast_fp16")]; bool matmul_331_transpose_x_0 = const()[name = string("matmul_331_transpose_x_0"), val = bool(false)]; bool matmul_331_transpose_y_0 = const()[name = string("matmul_331_transpose_y_0"), val = bool(false)]; tensor matmul_331_cast_fp16 = matmul(transpose_x = matmul_331_transpose_x_0, transpose_y = matmul_331_transpose_y_0, x = reshape_993_cast_fp16, y = reshape_994_cast_fp16)[name = string("matmul_331_cast_fp16")]; tensor concat_3319 = const()[name = string("concat_3319"), val = tensor([1, 1, 104, 104])]; tensor reshape_995_cast_fp16 = reshape(shape = concat_3319, x = matmul_331_cast_fp16)[name = string("reshape_995_cast_fp16")]; tensor transpose_3019_perm_0 = const()[name = string("transpose_3019_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3019 = transpose(perm = transpose_3019_perm_0, x = reshape_995_cast_fp16)[name = string("transpose_3228")]; tensor w_1327_cast_fp16 = add(x = transpose_3019, y = transpose_2305)[name = string("w_1327_cast_fp16")]; tensor var_6355_cast_fp16 = softmax(axis = var_6155, x = w_1327_cast_fp16)[name = string("op_6355_cast_fp16")]; string var_6357_equation_0 = const()[name = string("op_6357_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6357_cast_fp16 = einsum(equation = var_6357_equation_0, values = (var_6245_cast_fp16_11, var_6355_cast_fp16))[name = string("op_6357_cast_fp16")]; tensor transpose_664_perm_0 = const()[name = string("transpose_664_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3324 = const()[name = string("concat_3324"), val = tensor([1, 104, 64])]; tensor transpose_664_cast_fp16 = transpose(perm = transpose_664_perm_0, x = var_6211_cast_fp16_12)[name = string("transpose_3227")]; tensor reshape_996_cast_fp16 = reshape(shape = concat_3324, x = transpose_664_cast_fp16)[name = string("reshape_996_cast_fp16")]; tensor transpose_665_perm_0 = const()[name = string("transpose_665_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3325 = const()[name = string("concat_3325"), val = tensor([1, 64, 104])]; tensor transpose_665_cast_fp16 = transpose(perm = transpose_665_perm_0, x = var_6228_cast_fp16_12)[name = string("transpose_3226")]; tensor reshape_997_cast_fp16 = reshape(shape = concat_3325, x = transpose_665_cast_fp16)[name = string("reshape_997_cast_fp16")]; bool matmul_332_transpose_x_0 = const()[name = string("matmul_332_transpose_x_0"), val = bool(false)]; bool matmul_332_transpose_y_0 = const()[name = string("matmul_332_transpose_y_0"), val = bool(false)]; tensor matmul_332_cast_fp16 = matmul(transpose_x = matmul_332_transpose_x_0, transpose_y = matmul_332_transpose_y_0, x = reshape_996_cast_fp16, y = reshape_997_cast_fp16)[name = string("matmul_332_cast_fp16")]; tensor concat_3329 = const()[name = string("concat_3329"), val = tensor([1, 1, 104, 104])]; tensor reshape_998_cast_fp16 = reshape(shape = concat_3329, x = matmul_332_cast_fp16)[name = string("reshape_998_cast_fp16")]; tensor transpose_3020_perm_0 = const()[name = string("transpose_3020_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3020 = transpose(perm = transpose_3020_perm_0, x = reshape_998_cast_fp16)[name = string("transpose_3225")]; tensor w_1331_cast_fp16 = add(x = transpose_3020, y = transpose_2305)[name = string("w_1331_cast_fp16")]; tensor var_6363_cast_fp16 = softmax(axis = var_6155, x = w_1331_cast_fp16)[name = string("op_6363_cast_fp16")]; string var_6365_equation_0 = const()[name = string("op_6365_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6365_cast_fp16 = einsum(equation = var_6365_equation_0, values = (var_6245_cast_fp16_12, var_6363_cast_fp16))[name = string("op_6365_cast_fp16")]; tensor transpose_666_perm_0 = const()[name = string("transpose_666_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3334 = const()[name = string("concat_3334"), val = tensor([1, 104, 64])]; tensor transpose_666_cast_fp16 = transpose(perm = transpose_666_perm_0, x = var_6211_cast_fp16_13)[name = string("transpose_3224")]; tensor reshape_999_cast_fp16 = reshape(shape = concat_3334, x = transpose_666_cast_fp16)[name = string("reshape_999_cast_fp16")]; tensor transpose_667_perm_0 = const()[name = string("transpose_667_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3335 = const()[name = string("concat_3335"), val = tensor([1, 64, 104])]; tensor transpose_667_cast_fp16 = transpose(perm = transpose_667_perm_0, x = var_6228_cast_fp16_13)[name = string("transpose_3223")]; tensor reshape_1000_cast_fp16 = reshape(shape = concat_3335, x = transpose_667_cast_fp16)[name = string("reshape_1000_cast_fp16")]; bool matmul_333_transpose_x_0 = const()[name = string("matmul_333_transpose_x_0"), val = bool(false)]; bool matmul_333_transpose_y_0 = const()[name = string("matmul_333_transpose_y_0"), val = bool(false)]; tensor matmul_333_cast_fp16 = matmul(transpose_x = matmul_333_transpose_x_0, transpose_y = matmul_333_transpose_y_0, x = reshape_999_cast_fp16, y = reshape_1000_cast_fp16)[name = string("matmul_333_cast_fp16")]; tensor concat_3339 = const()[name = string("concat_3339"), val = tensor([1, 1, 104, 104])]; tensor reshape_1001_cast_fp16 = reshape(shape = concat_3339, x = matmul_333_cast_fp16)[name = string("reshape_1001_cast_fp16")]; tensor transpose_3021_perm_0 = const()[name = string("transpose_3021_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3021 = transpose(perm = transpose_3021_perm_0, x = reshape_1001_cast_fp16)[name = string("transpose_3222")]; tensor w_1335_cast_fp16 = add(x = transpose_3021, y = transpose_2305)[name = string("w_1335_cast_fp16")]; tensor var_6371_cast_fp16 = softmax(axis = var_6155, x = w_1335_cast_fp16)[name = string("op_6371_cast_fp16")]; string var_6373_equation_0 = const()[name = string("op_6373_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6373_cast_fp16 = einsum(equation = var_6373_equation_0, values = (var_6245_cast_fp16_13, var_6371_cast_fp16))[name = string("op_6373_cast_fp16")]; tensor transpose_668_perm_0 = const()[name = string("transpose_668_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3344 = const()[name = string("concat_3344"), val = tensor([1, 104, 64])]; tensor transpose_668_cast_fp16 = transpose(perm = transpose_668_perm_0, x = var_6211_cast_fp16_14)[name = string("transpose_3221")]; tensor reshape_1002_cast_fp16 = reshape(shape = concat_3344, x = transpose_668_cast_fp16)[name = string("reshape_1002_cast_fp16")]; tensor transpose_669_perm_0 = const()[name = string("transpose_669_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3345 = const()[name = string("concat_3345"), val = tensor([1, 64, 104])]; tensor transpose_669_cast_fp16 = transpose(perm = transpose_669_perm_0, x = var_6228_cast_fp16_14)[name = string("transpose_3220")]; tensor reshape_1003_cast_fp16 = reshape(shape = concat_3345, x = transpose_669_cast_fp16)[name = string("reshape_1003_cast_fp16")]; bool matmul_334_transpose_x_0 = const()[name = string("matmul_334_transpose_x_0"), val = bool(false)]; bool matmul_334_transpose_y_0 = const()[name = string("matmul_334_transpose_y_0"), val = bool(false)]; tensor matmul_334_cast_fp16 = matmul(transpose_x = matmul_334_transpose_x_0, transpose_y = matmul_334_transpose_y_0, x = reshape_1002_cast_fp16, y = reshape_1003_cast_fp16)[name = string("matmul_334_cast_fp16")]; tensor concat_3349 = const()[name = string("concat_3349"), val = tensor([1, 1, 104, 104])]; tensor reshape_1004_cast_fp16 = reshape(shape = concat_3349, x = matmul_334_cast_fp16)[name = string("reshape_1004_cast_fp16")]; tensor transpose_3022_perm_0 = const()[name = string("transpose_3022_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3022 = transpose(perm = transpose_3022_perm_0, x = reshape_1004_cast_fp16)[name = string("transpose_3219")]; tensor w_1339_cast_fp16 = add(x = transpose_3022, y = transpose_2305)[name = string("w_1339_cast_fp16")]; tensor var_6379_cast_fp16 = softmax(axis = var_6155, x = w_1339_cast_fp16)[name = string("op_6379_cast_fp16")]; string var_6381_equation_0 = const()[name = string("op_6381_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6381_cast_fp16 = einsum(equation = var_6381_equation_0, values = (var_6245_cast_fp16_14, var_6379_cast_fp16))[name = string("op_6381_cast_fp16")]; tensor transpose_670_perm_0 = const()[name = string("transpose_670_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3354 = const()[name = string("concat_3354"), val = tensor([1, 104, 64])]; tensor transpose_670_cast_fp16 = transpose(perm = transpose_670_perm_0, x = var_6211_cast_fp16_15)[name = string("transpose_3218")]; tensor reshape_1005_cast_fp16 = reshape(shape = concat_3354, x = transpose_670_cast_fp16)[name = string("reshape_1005_cast_fp16")]; tensor transpose_671_perm_0 = const()[name = string("transpose_671_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3355 = const()[name = string("concat_3355"), val = tensor([1, 64, 104])]; tensor transpose_671_cast_fp16 = transpose(perm = transpose_671_perm_0, x = var_6228_cast_fp16_15)[name = string("transpose_3217")]; tensor reshape_1006_cast_fp16 = reshape(shape = concat_3355, x = transpose_671_cast_fp16)[name = string("reshape_1006_cast_fp16")]; bool matmul_335_transpose_x_0 = const()[name = string("matmul_335_transpose_x_0"), val = bool(false)]; bool matmul_335_transpose_y_0 = const()[name = string("matmul_335_transpose_y_0"), val = bool(false)]; tensor matmul_335_cast_fp16 = matmul(transpose_x = matmul_335_transpose_x_0, transpose_y = matmul_335_transpose_y_0, x = reshape_1005_cast_fp16, y = reshape_1006_cast_fp16)[name = string("matmul_335_cast_fp16")]; tensor concat_3359 = const()[name = string("concat_3359"), val = tensor([1, 1, 104, 104])]; tensor reshape_1007_cast_fp16 = reshape(shape = concat_3359, x = matmul_335_cast_fp16)[name = string("reshape_1007_cast_fp16")]; tensor transpose_3023_perm_0 = const()[name = string("transpose_3023_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3023 = transpose(perm = transpose_3023_perm_0, x = reshape_1007_cast_fp16)[name = string("transpose_3216")]; tensor w_1343_cast_fp16 = add(x = transpose_3023, y = transpose_2305)[name = string("w_1343_cast_fp16")]; tensor var_6387_cast_fp16 = softmax(axis = var_6155, x = w_1343_cast_fp16)[name = string("op_6387_cast_fp16")]; string var_6389_equation_0 = const()[name = string("op_6389_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6389_cast_fp16 = einsum(equation = var_6389_equation_0, values = (var_6245_cast_fp16_15, var_6387_cast_fp16))[name = string("op_6389_cast_fp16")]; bool input_171_interleave_0 = const()[name = string("input_171_interleave_0"), val = bool(false)]; tensor input_171_cast_fp16 = concat(axis = var_6155, interleave = input_171_interleave_0, values = (var_6269_cast_fp16, var_6277_cast_fp16, var_6285_cast_fp16, var_6293_cast_fp16, var_6301_cast_fp16, var_6309_cast_fp16, var_6317_cast_fp16, var_6325_cast_fp16, var_6333_cast_fp16, var_6341_cast_fp16, var_6349_cast_fp16, var_6357_cast_fp16, var_6365_cast_fp16, var_6373_cast_fp16, var_6381_cast_fp16, var_6389_cast_fp16))[name = string("input_171_cast_fp16")]; string var_6398_pad_type_0 = const()[name = string("op_6398_pad_type_0"), val = string("valid")]; tensor var_6398_strides_0 = const()[name = string("op_6398_strides_0"), val = tensor([1, 1])]; tensor var_6398_pad_0 = const()[name = string("op_6398_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6398_dilations_0 = const()[name = string("op_6398_dilations_0"), val = tensor([1, 1])]; int32 var_6398_groups_0 = const()[name = string("op_6398_groups_0"), val = int32(1)]; tensor layers_20_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_20_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(534245696)))]; tensor layers_20_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_20_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(536342912)))]; tensor var_6398_cast_fp16 = conv(bias = layers_20_self_attn_out_proj_bias_to_fp16, dilations = var_6398_dilations_0, groups = var_6398_groups_0, pad = var_6398_pad_0, pad_type = var_6398_pad_type_0, strides = var_6398_strides_0, weight = layers_20_self_attn_out_proj_weight_to_fp16, x = input_171_cast_fp16)[name = string("op_6398_cast_fp16")]; tensor x_217_cast_fp16 = add(x = x_213_cast_fp16, y = var_6398_cast_fp16)[name = string("x_217_cast_fp16")]; tensor mu_83_axes_0 = const()[name = string("mu_83_axes_0"), val = tensor([1])]; bool mu_83_keep_dims_0 = const()[name = string("mu_83_keep_dims_0"), val = bool(true)]; tensor mu_83_cast_fp16 = reduce_mean(axes = mu_83_axes_0, keep_dims = mu_83_keep_dims_0, x = x_217_cast_fp16)[name = string("mu_83_cast_fp16")]; tensor var_6404_cast_fp16 = sub(x = x_217_cast_fp16, y = mu_83_cast_fp16)[name = string("op_6404_cast_fp16")]; fp16 var_6158_promoted_1_to_fp16 = const()[name = string("op_6158_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_6405_cast_fp16 = pow(x = var_6404_cast_fp16, y = var_6158_promoted_1_to_fp16)[name = string("op_6405_cast_fp16")]; tensor var_83_axes_0 = const()[name = string("var_83_axes_0"), val = tensor([1])]; bool var_83_keep_dims_0 = const()[name = string("var_83_keep_dims_0"), val = bool(true)]; tensor var_83_cast_fp16 = reduce_mean(axes = var_83_axes_0, keep_dims = var_83_keep_dims_0, x = var_6405_cast_fp16)[name = string("var_83_cast_fp16")]; fp16 var_6409_to_fp16 = const()[name = string("op_6409_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_6410_cast_fp16 = add(x = var_83_cast_fp16, y = var_6409_to_fp16)[name = string("op_6410_cast_fp16")]; fp32 var_6411_epsilon_0 = const()[name = string("op_6411_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_6411_cast_fp16 = rsqrt(epsilon = var_6411_epsilon_0, x = var_6410_cast_fp16)[name = string("op_6411_cast_fp16")]; tensor x_219_cast_fp16 = mul(x = var_6404_cast_fp16, y = var_6411_cast_fp16)[name = string("x_219_cast_fp16")]; tensor input_173_gamma_0_to_fp16 = const()[name = string("input_173_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(536345024)))]; tensor input_173_beta_0_to_fp16 = const()[name = string("input_173_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(536347136)))]; fp16 input_173_epsilon_0_to_fp16 = const()[name = string("input_173_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_173_cast_fp16 = batch_norm(beta = input_173_beta_0_to_fp16, epsilon = input_173_epsilon_0_to_fp16, gamma = input_173_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_219_cast_fp16)[name = string("input_173_cast_fp16")]; string x_221_pad_type_0 = const()[name = string("x_221_pad_type_0"), val = string("valid")]; tensor x_221_strides_0 = const()[name = string("x_221_strides_0"), val = tensor([1, 1])]; tensor x_221_pad_0 = const()[name = string("x_221_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_221_dilations_0 = const()[name = string("x_221_dilations_0"), val = tensor([1, 1])]; int32 x_221_groups_0 = const()[name = string("x_221_groups_0"), val = int32(1)]; tensor layers_20_fc1_weight_to_fp16 = const()[name = string("layers_20_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(536349248)))]; tensor layers_20_fc1_bias_to_fp16 = const()[name = string("layers_20_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(544737920)))]; tensor x_221_cast_fp16 = conv(bias = layers_20_fc1_bias_to_fp16, dilations = x_221_dilations_0, groups = x_221_groups_0, pad = x_221_pad_0, pad_type = x_221_pad_type_0, strides = x_221_strides_0, weight = layers_20_fc1_weight_to_fp16, x = input_173_cast_fp16)[name = string("x_221_cast_fp16")]; fp16 var_6426_to_fp16 = const()[name = string("op_6426_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_6427_cast_fp16 = mul(x = x_221_cast_fp16, y = var_6426_to_fp16)[name = string("op_6427_cast_fp16")]; tensor var_6428_cast_fp16 = mul(x = var_6427_cast_fp16, y = x_221_cast_fp16)[name = string("op_6428_cast_fp16")]; tensor var_6429_cast_fp16 = mul(x = var_6428_cast_fp16, y = x_221_cast_fp16)[name = string("op_6429_cast_fp16")]; tensor var_6430_cast_fp16 = add(x = x_221_cast_fp16, y = var_6429_cast_fp16)[name = string("op_6430_cast_fp16")]; fp16 var_6431_to_fp16 = const()[name = string("op_6431_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_47_cast_fp16 = mul(x = var_6430_cast_fp16, y = var_6431_to_fp16)[name = string("u_47_cast_fp16")]; fp16 var_6433_to_fp16 = const()[name = string("op_6433_to_fp16"), val = fp16(0x1p-1)]; tensor var_6434_cast_fp16 = mul(x = x_221_cast_fp16, y = var_6433_to_fp16)[name = string("op_6434_cast_fp16")]; tensor var_6435_cast_fp16 = tanh(x = u_47_cast_fp16)[name = string("op_6435_cast_fp16")]; fp16 var_6436_to_fp16 = const()[name = string("op_6436_to_fp16"), val = fp16(0x1p+0)]; tensor var_6437_cast_fp16 = add(x = var_6435_cast_fp16, y = var_6436_to_fp16)[name = string("op_6437_cast_fp16")]; tensor input_175_cast_fp16 = mul(x = var_6434_cast_fp16, y = var_6437_cast_fp16)[name = string("input_175_cast_fp16")]; string h_41_pad_type_0 = const()[name = string("h_41_pad_type_0"), val = string("valid")]; tensor h_41_strides_0 = const()[name = string("h_41_strides_0"), val = tensor([1, 1])]; tensor h_41_pad_0 = const()[name = string("h_41_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_41_dilations_0 = const()[name = string("h_41_dilations_0"), val = tensor([1, 1])]; int32 h_41_groups_0 = const()[name = string("h_41_groups_0"), val = int32(1)]; tensor layers_20_fc2_weight_to_fp16 = const()[name = string("layers_20_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(544746176)))]; tensor layers_20_fc2_bias_to_fp16 = const()[name = string("layers_20_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(553134848)))]; tensor h_41_cast_fp16 = conv(bias = layers_20_fc2_bias_to_fp16, dilations = h_41_dilations_0, groups = h_41_groups_0, pad = h_41_pad_0, pad_type = h_41_pad_type_0, strides = h_41_strides_0, weight = layers_20_fc2_weight_to_fp16, x = input_175_cast_fp16)[name = string("h_41_cast_fp16")]; tensor x_223_cast_fp16 = add(x = x_217_cast_fp16, y = h_41_cast_fp16)[name = string("x_223_cast_fp16")]; int32 var_6453 = const()[name = string("op_6453"), val = int32(1)]; tensor mu_85_axes_0 = const()[name = string("mu_85_axes_0"), val = tensor([1])]; bool mu_85_keep_dims_0 = const()[name = string("mu_85_keep_dims_0"), val = bool(true)]; tensor mu_85_cast_fp16 = reduce_mean(axes = mu_85_axes_0, keep_dims = mu_85_keep_dims_0, x = x_223_cast_fp16)[name = string("mu_85_cast_fp16")]; tensor var_6467_cast_fp16 = sub(x = x_223_cast_fp16, y = mu_85_cast_fp16)[name = string("op_6467_cast_fp16")]; fp16 var_6456_promoted_to_fp16 = const()[name = string("op_6456_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_6468_cast_fp16 = pow(x = var_6467_cast_fp16, y = var_6456_promoted_to_fp16)[name = string("op_6468_cast_fp16")]; tensor var_85_axes_0 = const()[name = string("var_85_axes_0"), val = tensor([1])]; bool var_85_keep_dims_0 = const()[name = string("var_85_keep_dims_0"), val = bool(true)]; tensor var_85_cast_fp16 = reduce_mean(axes = var_85_axes_0, keep_dims = var_85_keep_dims_0, x = var_6468_cast_fp16)[name = string("var_85_cast_fp16")]; fp16 var_6472_to_fp16 = const()[name = string("op_6472_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_6473_cast_fp16 = add(x = var_85_cast_fp16, y = var_6472_to_fp16)[name = string("op_6473_cast_fp16")]; fp32 var_6474_epsilon_0 = const()[name = string("op_6474_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_6474_cast_fp16 = rsqrt(epsilon = var_6474_epsilon_0, x = var_6473_cast_fp16)[name = string("op_6474_cast_fp16")]; tensor x_225_cast_fp16 = mul(x = var_6467_cast_fp16, y = var_6474_cast_fp16)[name = string("x_225_cast_fp16")]; tensor input_177_gamma_0_to_fp16 = const()[name = string("input_177_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(553136960)))]; tensor input_177_beta_0_to_fp16 = const()[name = string("input_177_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(553139072)))]; fp16 input_177_epsilon_0_to_fp16 = const()[name = string("input_177_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_177_cast_fp16 = batch_norm(beta = input_177_beta_0_to_fp16, epsilon = input_177_epsilon_0_to_fp16, gamma = input_177_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_225_cast_fp16)[name = string("input_177_cast_fp16")]; string var_6492_pad_type_0 = const()[name = string("op_6492_pad_type_0"), val = string("valid")]; tensor var_6492_strides_0 = const()[name = string("op_6492_strides_0"), val = tensor([1, 1])]; tensor var_6492_pad_0 = const()[name = string("op_6492_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6492_dilations_0 = const()[name = string("op_6492_dilations_0"), val = tensor([1, 1])]; int32 var_6492_groups_0 = const()[name = string("op_6492_groups_0"), val = int32(1)]; tensor var_6494_weight_0_to_fp16 = const()[name = string("op_6494_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(553141184)))]; tensor var_6494_bias_0_to_fp16 = const()[name = string("op_6494_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(555238400)))]; tensor var_6494_cast_fp16 = conv(bias = var_6494_bias_0_to_fp16, dilations = var_6492_dilations_0, groups = var_6492_groups_0, pad = var_6492_pad_0, pad_type = var_6492_pad_type_0, strides = var_6492_strides_0, weight = var_6494_weight_0_to_fp16, x = input_177_cast_fp16)[name = string("op_6494_cast_fp16")]; string var_6501_pad_type_0 = const()[name = string("op_6501_pad_type_0"), val = string("valid")]; tensor var_6501_strides_0 = const()[name = string("op_6501_strides_0"), val = tensor([1, 1])]; tensor var_6501_pad_0 = const()[name = string("op_6501_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6501_dilations_0 = const()[name = string("op_6501_dilations_0"), val = tensor([1, 1])]; int32 var_6501_groups_0 = const()[name = string("op_6501_groups_0"), val = int32(1)]; tensor layers_21_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_21_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(555240512)))]; tensor layers_21_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_21_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(557337728)))]; tensor var_6501_cast_fp16 = conv(bias = layers_21_self_attn_k_proj_bias_to_fp16, dilations = var_6501_dilations_0, groups = var_6501_groups_0, pad = var_6501_pad_0, pad_type = var_6501_pad_type_0, strides = var_6501_strides_0, weight = layers_21_self_attn_k_proj_weight_to_fp16, x = input_177_cast_fp16)[name = string("op_6501_cast_fp16")]; string var_6508_pad_type_0 = const()[name = string("op_6508_pad_type_0"), val = string("valid")]; tensor var_6508_strides_0 = const()[name = string("op_6508_strides_0"), val = tensor([1, 1])]; tensor var_6508_pad_0 = const()[name = string("op_6508_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6508_dilations_0 = const()[name = string("op_6508_dilations_0"), val = tensor([1, 1])]; int32 var_6508_groups_0 = const()[name = string("op_6508_groups_0"), val = int32(1)]; tensor layers_21_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_21_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(557339840)))]; tensor layers_21_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_21_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(559437056)))]; tensor var_6508_cast_fp16 = conv(bias = layers_21_self_attn_v_proj_bias_to_fp16, dilations = var_6508_dilations_0, groups = var_6508_groups_0, pad = var_6508_pad_0, pad_type = var_6508_pad_type_0, strides = var_6508_strides_0, weight = layers_21_self_attn_v_proj_weight_to_fp16, x = input_177_cast_fp16)[name = string("op_6508_cast_fp16")]; tensor tile_63 = const()[name = string("tile_63"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(559439168)))]; int32 var_6509_axis_0 = const()[name = string("op_6509_axis_0"), val = int32(1)]; tensor var_6509_cast_fp16_0, tensor var_6509_cast_fp16_1, tensor var_6509_cast_fp16_2, tensor var_6509_cast_fp16_3, tensor var_6509_cast_fp16_4, tensor var_6509_cast_fp16_5, tensor var_6509_cast_fp16_6, tensor var_6509_cast_fp16_7, tensor var_6509_cast_fp16_8, tensor var_6509_cast_fp16_9, tensor var_6509_cast_fp16_10, tensor var_6509_cast_fp16_11, tensor var_6509_cast_fp16_12, tensor var_6509_cast_fp16_13, tensor var_6509_cast_fp16_14, tensor var_6509_cast_fp16_15 = split(axis = var_6509_axis_0, split_sizes = tile_63, x = var_6494_cast_fp16)[name = string("op_6509_cast_fp16")]; tensor tile_64 = const()[name = string("tile_64"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(559439296)))]; int32 var_6526_axis_0 = const()[name = string("op_6526_axis_0"), val = int32(1)]; tensor var_6526_cast_fp16_0, tensor var_6526_cast_fp16_1, tensor var_6526_cast_fp16_2, tensor var_6526_cast_fp16_3, tensor var_6526_cast_fp16_4, tensor var_6526_cast_fp16_5, tensor var_6526_cast_fp16_6, tensor var_6526_cast_fp16_7, tensor var_6526_cast_fp16_8, tensor var_6526_cast_fp16_9, tensor var_6526_cast_fp16_10, tensor var_6526_cast_fp16_11, tensor var_6526_cast_fp16_12, tensor var_6526_cast_fp16_13, tensor var_6526_cast_fp16_14, tensor var_6526_cast_fp16_15 = split(axis = var_6526_axis_0, split_sizes = tile_64, x = var_6501_cast_fp16)[name = string("op_6526_cast_fp16")]; tensor tile_65 = const()[name = string("tile_65"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(559439424)))]; int32 var_6543_axis_0 = const()[name = string("op_6543_axis_0"), val = int32(1)]; tensor var_6543_cast_fp16_0, tensor var_6543_cast_fp16_1, tensor var_6543_cast_fp16_2, tensor var_6543_cast_fp16_3, tensor var_6543_cast_fp16_4, tensor var_6543_cast_fp16_5, tensor var_6543_cast_fp16_6, tensor var_6543_cast_fp16_7, tensor var_6543_cast_fp16_8, tensor var_6543_cast_fp16_9, tensor var_6543_cast_fp16_10, tensor var_6543_cast_fp16_11, tensor var_6543_cast_fp16_12, tensor var_6543_cast_fp16_13, tensor var_6543_cast_fp16_14, tensor var_6543_cast_fp16_15 = split(axis = var_6543_axis_0, split_sizes = tile_65, x = var_6508_cast_fp16)[name = string("op_6543_cast_fp16")]; tensor transpose_672_perm_0 = const()[name = string("transpose_672_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3364 = const()[name = string("concat_3364"), val = tensor([1, 104, 64])]; tensor transpose_672_cast_fp16 = transpose(perm = transpose_672_perm_0, x = var_6509_cast_fp16_0)[name = string("transpose_3215")]; tensor reshape_1008_cast_fp16 = reshape(shape = concat_3364, x = transpose_672_cast_fp16)[name = string("reshape_1008_cast_fp16")]; tensor transpose_673_perm_0 = const()[name = string("transpose_673_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3365 = const()[name = string("concat_3365"), val = tensor([1, 64, 104])]; tensor transpose_673_cast_fp16 = transpose(perm = transpose_673_perm_0, x = var_6526_cast_fp16_0)[name = string("transpose_3214")]; tensor reshape_1009_cast_fp16 = reshape(shape = concat_3365, x = transpose_673_cast_fp16)[name = string("reshape_1009_cast_fp16")]; bool matmul_336_transpose_x_0 = const()[name = string("matmul_336_transpose_x_0"), val = bool(false)]; bool matmul_336_transpose_y_0 = const()[name = string("matmul_336_transpose_y_0"), val = bool(false)]; tensor matmul_336_cast_fp16 = matmul(transpose_x = matmul_336_transpose_x_0, transpose_y = matmul_336_transpose_y_0, x = reshape_1008_cast_fp16, y = reshape_1009_cast_fp16)[name = string("matmul_336_cast_fp16")]; tensor concat_3369 = const()[name = string("concat_3369"), val = tensor([1, 1, 104, 104])]; tensor reshape_1010_cast_fp16 = reshape(shape = concat_3369, x = matmul_336_cast_fp16)[name = string("reshape_1010_cast_fp16")]; tensor transpose_3024_perm_0 = const()[name = string("transpose_3024_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3024 = transpose(perm = transpose_3024_perm_0, x = reshape_1010_cast_fp16)[name = string("transpose_3213")]; tensor w_1347_cast_fp16 = add(x = transpose_3024, y = transpose_2305)[name = string("w_1347_cast_fp16")]; tensor var_6565_cast_fp16 = softmax(axis = var_6453, x = w_1347_cast_fp16)[name = string("op_6565_cast_fp16")]; string var_6567_equation_0 = const()[name = string("op_6567_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6567_cast_fp16 = einsum(equation = var_6567_equation_0, values = (var_6543_cast_fp16_0, var_6565_cast_fp16))[name = string("op_6567_cast_fp16")]; tensor transpose_674_perm_0 = const()[name = string("transpose_674_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3374 = const()[name = string("concat_3374"), val = tensor([1, 104, 64])]; tensor transpose_674_cast_fp16 = transpose(perm = transpose_674_perm_0, x = var_6509_cast_fp16_1)[name = string("transpose_3212")]; tensor reshape_1011_cast_fp16 = reshape(shape = concat_3374, x = transpose_674_cast_fp16)[name = string("reshape_1011_cast_fp16")]; tensor transpose_675_perm_0 = const()[name = string("transpose_675_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3375 = const()[name = string("concat_3375"), val = tensor([1, 64, 104])]; tensor transpose_675_cast_fp16 = transpose(perm = transpose_675_perm_0, x = var_6526_cast_fp16_1)[name = string("transpose_3211")]; tensor reshape_1012_cast_fp16 = reshape(shape = concat_3375, x = transpose_675_cast_fp16)[name = string("reshape_1012_cast_fp16")]; bool matmul_337_transpose_x_0 = const()[name = string("matmul_337_transpose_x_0"), val = bool(false)]; bool matmul_337_transpose_y_0 = const()[name = string("matmul_337_transpose_y_0"), val = bool(false)]; tensor matmul_337_cast_fp16 = matmul(transpose_x = matmul_337_transpose_x_0, transpose_y = matmul_337_transpose_y_0, x = reshape_1011_cast_fp16, y = reshape_1012_cast_fp16)[name = string("matmul_337_cast_fp16")]; tensor concat_3379 = const()[name = string("concat_3379"), val = tensor([1, 1, 104, 104])]; tensor reshape_1013_cast_fp16 = reshape(shape = concat_3379, x = matmul_337_cast_fp16)[name = string("reshape_1013_cast_fp16")]; tensor transpose_3025_perm_0 = const()[name = string("transpose_3025_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3025 = transpose(perm = transpose_3025_perm_0, x = reshape_1013_cast_fp16)[name = string("transpose_3210")]; tensor w_1351_cast_fp16 = add(x = transpose_3025, y = transpose_2305)[name = string("w_1351_cast_fp16")]; tensor var_6573_cast_fp16 = softmax(axis = var_6453, x = w_1351_cast_fp16)[name = string("op_6573_cast_fp16")]; string var_6575_equation_0 = const()[name = string("op_6575_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6575_cast_fp16 = einsum(equation = var_6575_equation_0, values = (var_6543_cast_fp16_1, var_6573_cast_fp16))[name = string("op_6575_cast_fp16")]; tensor transpose_676_perm_0 = const()[name = string("transpose_676_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3384 = const()[name = string("concat_3384"), val = tensor([1, 104, 64])]; tensor transpose_676_cast_fp16 = transpose(perm = transpose_676_perm_0, x = var_6509_cast_fp16_2)[name = string("transpose_3209")]; tensor reshape_1014_cast_fp16 = reshape(shape = concat_3384, x = transpose_676_cast_fp16)[name = string("reshape_1014_cast_fp16")]; tensor transpose_677_perm_0 = const()[name = string("transpose_677_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3385 = const()[name = string("concat_3385"), val = tensor([1, 64, 104])]; tensor transpose_677_cast_fp16 = transpose(perm = transpose_677_perm_0, x = var_6526_cast_fp16_2)[name = string("transpose_3208")]; tensor reshape_1015_cast_fp16 = reshape(shape = concat_3385, x = transpose_677_cast_fp16)[name = string("reshape_1015_cast_fp16")]; bool matmul_338_transpose_x_0 = const()[name = string("matmul_338_transpose_x_0"), val = bool(false)]; bool matmul_338_transpose_y_0 = const()[name = string("matmul_338_transpose_y_0"), val = bool(false)]; tensor matmul_338_cast_fp16 = matmul(transpose_x = matmul_338_transpose_x_0, transpose_y = matmul_338_transpose_y_0, x = reshape_1014_cast_fp16, y = reshape_1015_cast_fp16)[name = string("matmul_338_cast_fp16")]; tensor concat_3389 = const()[name = string("concat_3389"), val = tensor([1, 1, 104, 104])]; tensor reshape_1016_cast_fp16 = reshape(shape = concat_3389, x = matmul_338_cast_fp16)[name = string("reshape_1016_cast_fp16")]; tensor transpose_3026_perm_0 = const()[name = string("transpose_3026_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3026 = transpose(perm = transpose_3026_perm_0, x = reshape_1016_cast_fp16)[name = string("transpose_3207")]; tensor w_1355_cast_fp16 = add(x = transpose_3026, y = transpose_2305)[name = string("w_1355_cast_fp16")]; tensor var_6581_cast_fp16 = softmax(axis = var_6453, x = w_1355_cast_fp16)[name = string("op_6581_cast_fp16")]; string var_6583_equation_0 = const()[name = string("op_6583_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6583_cast_fp16 = einsum(equation = var_6583_equation_0, values = (var_6543_cast_fp16_2, var_6581_cast_fp16))[name = string("op_6583_cast_fp16")]; tensor transpose_678_perm_0 = const()[name = string("transpose_678_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3394 = const()[name = string("concat_3394"), val = tensor([1, 104, 64])]; tensor transpose_678_cast_fp16 = transpose(perm = transpose_678_perm_0, x = var_6509_cast_fp16_3)[name = string("transpose_3206")]; tensor reshape_1017_cast_fp16 = reshape(shape = concat_3394, x = transpose_678_cast_fp16)[name = string("reshape_1017_cast_fp16")]; tensor transpose_679_perm_0 = const()[name = string("transpose_679_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3395 = const()[name = string("concat_3395"), val = tensor([1, 64, 104])]; tensor transpose_679_cast_fp16 = transpose(perm = transpose_679_perm_0, x = var_6526_cast_fp16_3)[name = string("transpose_3205")]; tensor reshape_1018_cast_fp16 = reshape(shape = concat_3395, x = transpose_679_cast_fp16)[name = string("reshape_1018_cast_fp16")]; bool matmul_339_transpose_x_0 = const()[name = string("matmul_339_transpose_x_0"), val = bool(false)]; bool matmul_339_transpose_y_0 = const()[name = string("matmul_339_transpose_y_0"), val = bool(false)]; tensor matmul_339_cast_fp16 = matmul(transpose_x = matmul_339_transpose_x_0, transpose_y = matmul_339_transpose_y_0, x = reshape_1017_cast_fp16, y = reshape_1018_cast_fp16)[name = string("matmul_339_cast_fp16")]; tensor concat_3399 = const()[name = string("concat_3399"), val = tensor([1, 1, 104, 104])]; tensor reshape_1019_cast_fp16 = reshape(shape = concat_3399, x = matmul_339_cast_fp16)[name = string("reshape_1019_cast_fp16")]; tensor transpose_3027_perm_0 = const()[name = string("transpose_3027_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3027 = transpose(perm = transpose_3027_perm_0, x = reshape_1019_cast_fp16)[name = string("transpose_3204")]; tensor w_1359_cast_fp16 = add(x = transpose_3027, y = transpose_2305)[name = string("w_1359_cast_fp16")]; tensor var_6589_cast_fp16 = softmax(axis = var_6453, x = w_1359_cast_fp16)[name = string("op_6589_cast_fp16")]; string var_6591_equation_0 = const()[name = string("op_6591_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6591_cast_fp16 = einsum(equation = var_6591_equation_0, values = (var_6543_cast_fp16_3, var_6589_cast_fp16))[name = string("op_6591_cast_fp16")]; tensor transpose_680_perm_0 = const()[name = string("transpose_680_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3404 = const()[name = string("concat_3404"), val = tensor([1, 104, 64])]; tensor transpose_680_cast_fp16 = transpose(perm = transpose_680_perm_0, x = var_6509_cast_fp16_4)[name = string("transpose_3203")]; tensor reshape_1020_cast_fp16 = reshape(shape = concat_3404, x = transpose_680_cast_fp16)[name = string("reshape_1020_cast_fp16")]; tensor transpose_681_perm_0 = const()[name = string("transpose_681_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3405 = const()[name = string("concat_3405"), val = tensor([1, 64, 104])]; tensor transpose_681_cast_fp16 = transpose(perm = transpose_681_perm_0, x = var_6526_cast_fp16_4)[name = string("transpose_3202")]; tensor reshape_1021_cast_fp16 = reshape(shape = concat_3405, x = transpose_681_cast_fp16)[name = string("reshape_1021_cast_fp16")]; bool matmul_340_transpose_x_0 = const()[name = string("matmul_340_transpose_x_0"), val = bool(false)]; bool matmul_340_transpose_y_0 = const()[name = string("matmul_340_transpose_y_0"), val = bool(false)]; tensor matmul_340_cast_fp16 = matmul(transpose_x = matmul_340_transpose_x_0, transpose_y = matmul_340_transpose_y_0, x = reshape_1020_cast_fp16, y = reshape_1021_cast_fp16)[name = string("matmul_340_cast_fp16")]; tensor concat_3409 = const()[name = string("concat_3409"), val = tensor([1, 1, 104, 104])]; tensor reshape_1022_cast_fp16 = reshape(shape = concat_3409, x = matmul_340_cast_fp16)[name = string("reshape_1022_cast_fp16")]; tensor transpose_3028_perm_0 = const()[name = string("transpose_3028_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3028 = transpose(perm = transpose_3028_perm_0, x = reshape_1022_cast_fp16)[name = string("transpose_3201")]; tensor w_1363_cast_fp16 = add(x = transpose_3028, y = transpose_2305)[name = string("w_1363_cast_fp16")]; tensor var_6597_cast_fp16 = softmax(axis = var_6453, x = w_1363_cast_fp16)[name = string("op_6597_cast_fp16")]; string var_6599_equation_0 = const()[name = string("op_6599_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6599_cast_fp16 = einsum(equation = var_6599_equation_0, values = (var_6543_cast_fp16_4, var_6597_cast_fp16))[name = string("op_6599_cast_fp16")]; tensor transpose_682_perm_0 = const()[name = string("transpose_682_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3414 = const()[name = string("concat_3414"), val = tensor([1, 104, 64])]; tensor transpose_682_cast_fp16 = transpose(perm = transpose_682_perm_0, x = var_6509_cast_fp16_5)[name = string("transpose_3200")]; tensor reshape_1023_cast_fp16 = reshape(shape = concat_3414, x = transpose_682_cast_fp16)[name = string("reshape_1023_cast_fp16")]; tensor transpose_683_perm_0 = const()[name = string("transpose_683_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3415 = const()[name = string("concat_3415"), val = tensor([1, 64, 104])]; tensor transpose_683_cast_fp16 = transpose(perm = transpose_683_perm_0, x = var_6526_cast_fp16_5)[name = string("transpose_3199")]; tensor reshape_1024_cast_fp16 = reshape(shape = concat_3415, x = transpose_683_cast_fp16)[name = string("reshape_1024_cast_fp16")]; bool matmul_341_transpose_x_0 = const()[name = string("matmul_341_transpose_x_0"), val = bool(false)]; bool matmul_341_transpose_y_0 = const()[name = string("matmul_341_transpose_y_0"), val = bool(false)]; tensor matmul_341_cast_fp16 = matmul(transpose_x = matmul_341_transpose_x_0, transpose_y = matmul_341_transpose_y_0, x = reshape_1023_cast_fp16, y = reshape_1024_cast_fp16)[name = string("matmul_341_cast_fp16")]; tensor concat_3419 = const()[name = string("concat_3419"), val = tensor([1, 1, 104, 104])]; tensor reshape_1025_cast_fp16 = reshape(shape = concat_3419, x = matmul_341_cast_fp16)[name = string("reshape_1025_cast_fp16")]; tensor transpose_3029_perm_0 = const()[name = string("transpose_3029_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3029 = transpose(perm = transpose_3029_perm_0, x = reshape_1025_cast_fp16)[name = string("transpose_3198")]; tensor w_1367_cast_fp16 = add(x = transpose_3029, y = transpose_2305)[name = string("w_1367_cast_fp16")]; tensor var_6605_cast_fp16 = softmax(axis = var_6453, x = w_1367_cast_fp16)[name = string("op_6605_cast_fp16")]; string var_6607_equation_0 = const()[name = string("op_6607_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6607_cast_fp16 = einsum(equation = var_6607_equation_0, values = (var_6543_cast_fp16_5, var_6605_cast_fp16))[name = string("op_6607_cast_fp16")]; tensor transpose_684_perm_0 = const()[name = string("transpose_684_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3424 = const()[name = string("concat_3424"), val = tensor([1, 104, 64])]; tensor transpose_684_cast_fp16 = transpose(perm = transpose_684_perm_0, x = var_6509_cast_fp16_6)[name = string("transpose_3197")]; tensor reshape_1026_cast_fp16 = reshape(shape = concat_3424, x = transpose_684_cast_fp16)[name = string("reshape_1026_cast_fp16")]; tensor transpose_685_perm_0 = const()[name = string("transpose_685_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3425 = const()[name = string("concat_3425"), val = tensor([1, 64, 104])]; tensor transpose_685_cast_fp16 = transpose(perm = transpose_685_perm_0, x = var_6526_cast_fp16_6)[name = string("transpose_3196")]; tensor reshape_1027_cast_fp16 = reshape(shape = concat_3425, x = transpose_685_cast_fp16)[name = string("reshape_1027_cast_fp16")]; bool matmul_342_transpose_x_0 = const()[name = string("matmul_342_transpose_x_0"), val = bool(false)]; bool matmul_342_transpose_y_0 = const()[name = string("matmul_342_transpose_y_0"), val = bool(false)]; tensor matmul_342_cast_fp16 = matmul(transpose_x = matmul_342_transpose_x_0, transpose_y = matmul_342_transpose_y_0, x = reshape_1026_cast_fp16, y = reshape_1027_cast_fp16)[name = string("matmul_342_cast_fp16")]; tensor concat_3429 = const()[name = string("concat_3429"), val = tensor([1, 1, 104, 104])]; tensor reshape_1028_cast_fp16 = reshape(shape = concat_3429, x = matmul_342_cast_fp16)[name = string("reshape_1028_cast_fp16")]; tensor transpose_3030_perm_0 = const()[name = string("transpose_3030_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3030 = transpose(perm = transpose_3030_perm_0, x = reshape_1028_cast_fp16)[name = string("transpose_3195")]; tensor w_1371_cast_fp16 = add(x = transpose_3030, y = transpose_2305)[name = string("w_1371_cast_fp16")]; tensor var_6613_cast_fp16 = softmax(axis = var_6453, x = w_1371_cast_fp16)[name = string("op_6613_cast_fp16")]; string var_6615_equation_0 = const()[name = string("op_6615_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6615_cast_fp16 = einsum(equation = var_6615_equation_0, values = (var_6543_cast_fp16_6, var_6613_cast_fp16))[name = string("op_6615_cast_fp16")]; tensor transpose_686_perm_0 = const()[name = string("transpose_686_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3434 = const()[name = string("concat_3434"), val = tensor([1, 104, 64])]; tensor transpose_686_cast_fp16 = transpose(perm = transpose_686_perm_0, x = var_6509_cast_fp16_7)[name = string("transpose_3194")]; tensor reshape_1029_cast_fp16 = reshape(shape = concat_3434, x = transpose_686_cast_fp16)[name = string("reshape_1029_cast_fp16")]; tensor transpose_687_perm_0 = const()[name = string("transpose_687_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3435 = const()[name = string("concat_3435"), val = tensor([1, 64, 104])]; tensor transpose_687_cast_fp16 = transpose(perm = transpose_687_perm_0, x = var_6526_cast_fp16_7)[name = string("transpose_3193")]; tensor reshape_1030_cast_fp16 = reshape(shape = concat_3435, x = transpose_687_cast_fp16)[name = string("reshape_1030_cast_fp16")]; bool matmul_343_transpose_x_0 = const()[name = string("matmul_343_transpose_x_0"), val = bool(false)]; bool matmul_343_transpose_y_0 = const()[name = string("matmul_343_transpose_y_0"), val = bool(false)]; tensor matmul_343_cast_fp16 = matmul(transpose_x = matmul_343_transpose_x_0, transpose_y = matmul_343_transpose_y_0, x = reshape_1029_cast_fp16, y = reshape_1030_cast_fp16)[name = string("matmul_343_cast_fp16")]; tensor concat_3439 = const()[name = string("concat_3439"), val = tensor([1, 1, 104, 104])]; tensor reshape_1031_cast_fp16 = reshape(shape = concat_3439, x = matmul_343_cast_fp16)[name = string("reshape_1031_cast_fp16")]; tensor transpose_3031_perm_0 = const()[name = string("transpose_3031_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3031 = transpose(perm = transpose_3031_perm_0, x = reshape_1031_cast_fp16)[name = string("transpose_3192")]; tensor w_1375_cast_fp16 = add(x = transpose_3031, y = transpose_2305)[name = string("w_1375_cast_fp16")]; tensor var_6621_cast_fp16 = softmax(axis = var_6453, x = w_1375_cast_fp16)[name = string("op_6621_cast_fp16")]; string var_6623_equation_0 = const()[name = string("op_6623_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6623_cast_fp16 = einsum(equation = var_6623_equation_0, values = (var_6543_cast_fp16_7, var_6621_cast_fp16))[name = string("op_6623_cast_fp16")]; tensor transpose_688_perm_0 = const()[name = string("transpose_688_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3444 = const()[name = string("concat_3444"), val = tensor([1, 104, 64])]; tensor transpose_688_cast_fp16 = transpose(perm = transpose_688_perm_0, x = var_6509_cast_fp16_8)[name = string("transpose_3191")]; tensor reshape_1032_cast_fp16 = reshape(shape = concat_3444, x = transpose_688_cast_fp16)[name = string("reshape_1032_cast_fp16")]; tensor transpose_689_perm_0 = const()[name = string("transpose_689_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3445 = const()[name = string("concat_3445"), val = tensor([1, 64, 104])]; tensor transpose_689_cast_fp16 = transpose(perm = transpose_689_perm_0, x = var_6526_cast_fp16_8)[name = string("transpose_3190")]; tensor reshape_1033_cast_fp16 = reshape(shape = concat_3445, x = transpose_689_cast_fp16)[name = string("reshape_1033_cast_fp16")]; bool matmul_344_transpose_x_0 = const()[name = string("matmul_344_transpose_x_0"), val = bool(false)]; bool matmul_344_transpose_y_0 = const()[name = string("matmul_344_transpose_y_0"), val = bool(false)]; tensor matmul_344_cast_fp16 = matmul(transpose_x = matmul_344_transpose_x_0, transpose_y = matmul_344_transpose_y_0, x = reshape_1032_cast_fp16, y = reshape_1033_cast_fp16)[name = string("matmul_344_cast_fp16")]; tensor concat_3449 = const()[name = string("concat_3449"), val = tensor([1, 1, 104, 104])]; tensor reshape_1034_cast_fp16 = reshape(shape = concat_3449, x = matmul_344_cast_fp16)[name = string("reshape_1034_cast_fp16")]; tensor transpose_3032_perm_0 = const()[name = string("transpose_3032_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3032 = transpose(perm = transpose_3032_perm_0, x = reshape_1034_cast_fp16)[name = string("transpose_3189")]; tensor w_1379_cast_fp16 = add(x = transpose_3032, y = transpose_2305)[name = string("w_1379_cast_fp16")]; tensor var_6629_cast_fp16 = softmax(axis = var_6453, x = w_1379_cast_fp16)[name = string("op_6629_cast_fp16")]; string var_6631_equation_0 = const()[name = string("op_6631_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6631_cast_fp16 = einsum(equation = var_6631_equation_0, values = (var_6543_cast_fp16_8, var_6629_cast_fp16))[name = string("op_6631_cast_fp16")]; tensor transpose_690_perm_0 = const()[name = string("transpose_690_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3454 = const()[name = string("concat_3454"), val = tensor([1, 104, 64])]; tensor transpose_690_cast_fp16 = transpose(perm = transpose_690_perm_0, x = var_6509_cast_fp16_9)[name = string("transpose_3188")]; tensor reshape_1035_cast_fp16 = reshape(shape = concat_3454, x = transpose_690_cast_fp16)[name = string("reshape_1035_cast_fp16")]; tensor transpose_691_perm_0 = const()[name = string("transpose_691_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3455 = const()[name = string("concat_3455"), val = tensor([1, 64, 104])]; tensor transpose_691_cast_fp16 = transpose(perm = transpose_691_perm_0, x = var_6526_cast_fp16_9)[name = string("transpose_3187")]; tensor reshape_1036_cast_fp16 = reshape(shape = concat_3455, x = transpose_691_cast_fp16)[name = string("reshape_1036_cast_fp16")]; bool matmul_345_transpose_x_0 = const()[name = string("matmul_345_transpose_x_0"), val = bool(false)]; bool matmul_345_transpose_y_0 = const()[name = string("matmul_345_transpose_y_0"), val = bool(false)]; tensor matmul_345_cast_fp16 = matmul(transpose_x = matmul_345_transpose_x_0, transpose_y = matmul_345_transpose_y_0, x = reshape_1035_cast_fp16, y = reshape_1036_cast_fp16)[name = string("matmul_345_cast_fp16")]; tensor concat_3459 = const()[name = string("concat_3459"), val = tensor([1, 1, 104, 104])]; tensor reshape_1037_cast_fp16 = reshape(shape = concat_3459, x = matmul_345_cast_fp16)[name = string("reshape_1037_cast_fp16")]; tensor transpose_3033_perm_0 = const()[name = string("transpose_3033_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3033 = transpose(perm = transpose_3033_perm_0, x = reshape_1037_cast_fp16)[name = string("transpose_3186")]; tensor w_1383_cast_fp16 = add(x = transpose_3033, y = transpose_2305)[name = string("w_1383_cast_fp16")]; tensor var_6637_cast_fp16 = softmax(axis = var_6453, x = w_1383_cast_fp16)[name = string("op_6637_cast_fp16")]; string var_6639_equation_0 = const()[name = string("op_6639_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6639_cast_fp16 = einsum(equation = var_6639_equation_0, values = (var_6543_cast_fp16_9, var_6637_cast_fp16))[name = string("op_6639_cast_fp16")]; tensor transpose_692_perm_0 = const()[name = string("transpose_692_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3464 = const()[name = string("concat_3464"), val = tensor([1, 104, 64])]; tensor transpose_692_cast_fp16 = transpose(perm = transpose_692_perm_0, x = var_6509_cast_fp16_10)[name = string("transpose_3185")]; tensor reshape_1038_cast_fp16 = reshape(shape = concat_3464, x = transpose_692_cast_fp16)[name = string("reshape_1038_cast_fp16")]; tensor transpose_693_perm_0 = const()[name = string("transpose_693_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3465 = const()[name = string("concat_3465"), val = tensor([1, 64, 104])]; tensor transpose_693_cast_fp16 = transpose(perm = transpose_693_perm_0, x = var_6526_cast_fp16_10)[name = string("transpose_3184")]; tensor reshape_1039_cast_fp16 = reshape(shape = concat_3465, x = transpose_693_cast_fp16)[name = string("reshape_1039_cast_fp16")]; bool matmul_346_transpose_x_0 = const()[name = string("matmul_346_transpose_x_0"), val = bool(false)]; bool matmul_346_transpose_y_0 = const()[name = string("matmul_346_transpose_y_0"), val = bool(false)]; tensor matmul_346_cast_fp16 = matmul(transpose_x = matmul_346_transpose_x_0, transpose_y = matmul_346_transpose_y_0, x = reshape_1038_cast_fp16, y = reshape_1039_cast_fp16)[name = string("matmul_346_cast_fp16")]; tensor concat_3469 = const()[name = string("concat_3469"), val = tensor([1, 1, 104, 104])]; tensor reshape_1040_cast_fp16 = reshape(shape = concat_3469, x = matmul_346_cast_fp16)[name = string("reshape_1040_cast_fp16")]; tensor transpose_3034_perm_0 = const()[name = string("transpose_3034_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3034 = transpose(perm = transpose_3034_perm_0, x = reshape_1040_cast_fp16)[name = string("transpose_3183")]; tensor w_1387_cast_fp16 = add(x = transpose_3034, y = transpose_2305)[name = string("w_1387_cast_fp16")]; tensor var_6645_cast_fp16 = softmax(axis = var_6453, x = w_1387_cast_fp16)[name = string("op_6645_cast_fp16")]; string var_6647_equation_0 = const()[name = string("op_6647_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6647_cast_fp16 = einsum(equation = var_6647_equation_0, values = (var_6543_cast_fp16_10, var_6645_cast_fp16))[name = string("op_6647_cast_fp16")]; tensor transpose_694_perm_0 = const()[name = string("transpose_694_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3474 = const()[name = string("concat_3474"), val = tensor([1, 104, 64])]; tensor transpose_694_cast_fp16 = transpose(perm = transpose_694_perm_0, x = var_6509_cast_fp16_11)[name = string("transpose_3182")]; tensor reshape_1041_cast_fp16 = reshape(shape = concat_3474, x = transpose_694_cast_fp16)[name = string("reshape_1041_cast_fp16")]; tensor transpose_695_perm_0 = const()[name = string("transpose_695_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3475 = const()[name = string("concat_3475"), val = tensor([1, 64, 104])]; tensor transpose_695_cast_fp16 = transpose(perm = transpose_695_perm_0, x = var_6526_cast_fp16_11)[name = string("transpose_3181")]; tensor reshape_1042_cast_fp16 = reshape(shape = concat_3475, x = transpose_695_cast_fp16)[name = string("reshape_1042_cast_fp16")]; bool matmul_347_transpose_x_0 = const()[name = string("matmul_347_transpose_x_0"), val = bool(false)]; bool matmul_347_transpose_y_0 = const()[name = string("matmul_347_transpose_y_0"), val = bool(false)]; tensor matmul_347_cast_fp16 = matmul(transpose_x = matmul_347_transpose_x_0, transpose_y = matmul_347_transpose_y_0, x = reshape_1041_cast_fp16, y = reshape_1042_cast_fp16)[name = string("matmul_347_cast_fp16")]; tensor concat_3479 = const()[name = string("concat_3479"), val = tensor([1, 1, 104, 104])]; tensor reshape_1043_cast_fp16 = reshape(shape = concat_3479, x = matmul_347_cast_fp16)[name = string("reshape_1043_cast_fp16")]; tensor transpose_3035_perm_0 = const()[name = string("transpose_3035_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3035 = transpose(perm = transpose_3035_perm_0, x = reshape_1043_cast_fp16)[name = string("transpose_3180")]; tensor w_1391_cast_fp16 = add(x = transpose_3035, y = transpose_2305)[name = string("w_1391_cast_fp16")]; tensor var_6653_cast_fp16 = softmax(axis = var_6453, x = w_1391_cast_fp16)[name = string("op_6653_cast_fp16")]; string var_6655_equation_0 = const()[name = string("op_6655_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6655_cast_fp16 = einsum(equation = var_6655_equation_0, values = (var_6543_cast_fp16_11, var_6653_cast_fp16))[name = string("op_6655_cast_fp16")]; tensor transpose_696_perm_0 = const()[name = string("transpose_696_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3484 = const()[name = string("concat_3484"), val = tensor([1, 104, 64])]; tensor transpose_696_cast_fp16 = transpose(perm = transpose_696_perm_0, x = var_6509_cast_fp16_12)[name = string("transpose_3179")]; tensor reshape_1044_cast_fp16 = reshape(shape = concat_3484, x = transpose_696_cast_fp16)[name = string("reshape_1044_cast_fp16")]; tensor transpose_697_perm_0 = const()[name = string("transpose_697_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3485 = const()[name = string("concat_3485"), val = tensor([1, 64, 104])]; tensor transpose_697_cast_fp16 = transpose(perm = transpose_697_perm_0, x = var_6526_cast_fp16_12)[name = string("transpose_3178")]; tensor reshape_1045_cast_fp16 = reshape(shape = concat_3485, x = transpose_697_cast_fp16)[name = string("reshape_1045_cast_fp16")]; bool matmul_348_transpose_x_0 = const()[name = string("matmul_348_transpose_x_0"), val = bool(false)]; bool matmul_348_transpose_y_0 = const()[name = string("matmul_348_transpose_y_0"), val = bool(false)]; tensor matmul_348_cast_fp16 = matmul(transpose_x = matmul_348_transpose_x_0, transpose_y = matmul_348_transpose_y_0, x = reshape_1044_cast_fp16, y = reshape_1045_cast_fp16)[name = string("matmul_348_cast_fp16")]; tensor concat_3489 = const()[name = string("concat_3489"), val = tensor([1, 1, 104, 104])]; tensor reshape_1046_cast_fp16 = reshape(shape = concat_3489, x = matmul_348_cast_fp16)[name = string("reshape_1046_cast_fp16")]; tensor transpose_3036_perm_0 = const()[name = string("transpose_3036_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3036 = transpose(perm = transpose_3036_perm_0, x = reshape_1046_cast_fp16)[name = string("transpose_3177")]; tensor w_1395_cast_fp16 = add(x = transpose_3036, y = transpose_2305)[name = string("w_1395_cast_fp16")]; tensor var_6661_cast_fp16 = softmax(axis = var_6453, x = w_1395_cast_fp16)[name = string("op_6661_cast_fp16")]; string var_6663_equation_0 = const()[name = string("op_6663_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6663_cast_fp16 = einsum(equation = var_6663_equation_0, values = (var_6543_cast_fp16_12, var_6661_cast_fp16))[name = string("op_6663_cast_fp16")]; tensor transpose_698_perm_0 = const()[name = string("transpose_698_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3494 = const()[name = string("concat_3494"), val = tensor([1, 104, 64])]; tensor transpose_698_cast_fp16 = transpose(perm = transpose_698_perm_0, x = var_6509_cast_fp16_13)[name = string("transpose_3176")]; tensor reshape_1047_cast_fp16 = reshape(shape = concat_3494, x = transpose_698_cast_fp16)[name = string("reshape_1047_cast_fp16")]; tensor transpose_699_perm_0 = const()[name = string("transpose_699_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3495 = const()[name = string("concat_3495"), val = tensor([1, 64, 104])]; tensor transpose_699_cast_fp16 = transpose(perm = transpose_699_perm_0, x = var_6526_cast_fp16_13)[name = string("transpose_3175")]; tensor reshape_1048_cast_fp16 = reshape(shape = concat_3495, x = transpose_699_cast_fp16)[name = string("reshape_1048_cast_fp16")]; bool matmul_349_transpose_x_0 = const()[name = string("matmul_349_transpose_x_0"), val = bool(false)]; bool matmul_349_transpose_y_0 = const()[name = string("matmul_349_transpose_y_0"), val = bool(false)]; tensor matmul_349_cast_fp16 = matmul(transpose_x = matmul_349_transpose_x_0, transpose_y = matmul_349_transpose_y_0, x = reshape_1047_cast_fp16, y = reshape_1048_cast_fp16)[name = string("matmul_349_cast_fp16")]; tensor concat_3499 = const()[name = string("concat_3499"), val = tensor([1, 1, 104, 104])]; tensor reshape_1049_cast_fp16 = reshape(shape = concat_3499, x = matmul_349_cast_fp16)[name = string("reshape_1049_cast_fp16")]; tensor transpose_3037_perm_0 = const()[name = string("transpose_3037_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3037 = transpose(perm = transpose_3037_perm_0, x = reshape_1049_cast_fp16)[name = string("transpose_3174")]; tensor w_1399_cast_fp16 = add(x = transpose_3037, y = transpose_2305)[name = string("w_1399_cast_fp16")]; tensor var_6669_cast_fp16 = softmax(axis = var_6453, x = w_1399_cast_fp16)[name = string("op_6669_cast_fp16")]; string var_6671_equation_0 = const()[name = string("op_6671_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6671_cast_fp16 = einsum(equation = var_6671_equation_0, values = (var_6543_cast_fp16_13, var_6669_cast_fp16))[name = string("op_6671_cast_fp16")]; tensor transpose_700_perm_0 = const()[name = string("transpose_700_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3504 = const()[name = string("concat_3504"), val = tensor([1, 104, 64])]; tensor transpose_700_cast_fp16 = transpose(perm = transpose_700_perm_0, x = var_6509_cast_fp16_14)[name = string("transpose_3173")]; tensor reshape_1050_cast_fp16 = reshape(shape = concat_3504, x = transpose_700_cast_fp16)[name = string("reshape_1050_cast_fp16")]; tensor transpose_701_perm_0 = const()[name = string("transpose_701_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3505 = const()[name = string("concat_3505"), val = tensor([1, 64, 104])]; tensor transpose_701_cast_fp16 = transpose(perm = transpose_701_perm_0, x = var_6526_cast_fp16_14)[name = string("transpose_3172")]; tensor reshape_1051_cast_fp16 = reshape(shape = concat_3505, x = transpose_701_cast_fp16)[name = string("reshape_1051_cast_fp16")]; bool matmul_350_transpose_x_0 = const()[name = string("matmul_350_transpose_x_0"), val = bool(false)]; bool matmul_350_transpose_y_0 = const()[name = string("matmul_350_transpose_y_0"), val = bool(false)]; tensor matmul_350_cast_fp16 = matmul(transpose_x = matmul_350_transpose_x_0, transpose_y = matmul_350_transpose_y_0, x = reshape_1050_cast_fp16, y = reshape_1051_cast_fp16)[name = string("matmul_350_cast_fp16")]; tensor concat_3509 = const()[name = string("concat_3509"), val = tensor([1, 1, 104, 104])]; tensor reshape_1052_cast_fp16 = reshape(shape = concat_3509, x = matmul_350_cast_fp16)[name = string("reshape_1052_cast_fp16")]; tensor transpose_3038_perm_0 = const()[name = string("transpose_3038_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3038 = transpose(perm = transpose_3038_perm_0, x = reshape_1052_cast_fp16)[name = string("transpose_3171")]; tensor w_1403_cast_fp16 = add(x = transpose_3038, y = transpose_2305)[name = string("w_1403_cast_fp16")]; tensor var_6677_cast_fp16 = softmax(axis = var_6453, x = w_1403_cast_fp16)[name = string("op_6677_cast_fp16")]; string var_6679_equation_0 = const()[name = string("op_6679_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6679_cast_fp16 = einsum(equation = var_6679_equation_0, values = (var_6543_cast_fp16_14, var_6677_cast_fp16))[name = string("op_6679_cast_fp16")]; tensor transpose_702_perm_0 = const()[name = string("transpose_702_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3514 = const()[name = string("concat_3514"), val = tensor([1, 104, 64])]; tensor transpose_702_cast_fp16 = transpose(perm = transpose_702_perm_0, x = var_6509_cast_fp16_15)[name = string("transpose_3170")]; tensor reshape_1053_cast_fp16 = reshape(shape = concat_3514, x = transpose_702_cast_fp16)[name = string("reshape_1053_cast_fp16")]; tensor transpose_703_perm_0 = const()[name = string("transpose_703_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3515 = const()[name = string("concat_3515"), val = tensor([1, 64, 104])]; tensor transpose_703_cast_fp16 = transpose(perm = transpose_703_perm_0, x = var_6526_cast_fp16_15)[name = string("transpose_3169")]; tensor reshape_1054_cast_fp16 = reshape(shape = concat_3515, x = transpose_703_cast_fp16)[name = string("reshape_1054_cast_fp16")]; bool matmul_351_transpose_x_0 = const()[name = string("matmul_351_transpose_x_0"), val = bool(false)]; bool matmul_351_transpose_y_0 = const()[name = string("matmul_351_transpose_y_0"), val = bool(false)]; tensor matmul_351_cast_fp16 = matmul(transpose_x = matmul_351_transpose_x_0, transpose_y = matmul_351_transpose_y_0, x = reshape_1053_cast_fp16, y = reshape_1054_cast_fp16)[name = string("matmul_351_cast_fp16")]; tensor concat_3519 = const()[name = string("concat_3519"), val = tensor([1, 1, 104, 104])]; tensor reshape_1055_cast_fp16 = reshape(shape = concat_3519, x = matmul_351_cast_fp16)[name = string("reshape_1055_cast_fp16")]; tensor transpose_3039_perm_0 = const()[name = string("transpose_3039_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3039 = transpose(perm = transpose_3039_perm_0, x = reshape_1055_cast_fp16)[name = string("transpose_3168")]; tensor w_1407_cast_fp16 = add(x = transpose_3039, y = transpose_2305)[name = string("w_1407_cast_fp16")]; tensor var_6685_cast_fp16 = softmax(axis = var_6453, x = w_1407_cast_fp16)[name = string("op_6685_cast_fp16")]; string var_6687_equation_0 = const()[name = string("op_6687_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6687_cast_fp16 = einsum(equation = var_6687_equation_0, values = (var_6543_cast_fp16_15, var_6685_cast_fp16))[name = string("op_6687_cast_fp16")]; bool input_179_interleave_0 = const()[name = string("input_179_interleave_0"), val = bool(false)]; tensor input_179_cast_fp16 = concat(axis = var_6453, interleave = input_179_interleave_0, values = (var_6567_cast_fp16, var_6575_cast_fp16, var_6583_cast_fp16, var_6591_cast_fp16, var_6599_cast_fp16, var_6607_cast_fp16, var_6615_cast_fp16, var_6623_cast_fp16, var_6631_cast_fp16, var_6639_cast_fp16, var_6647_cast_fp16, var_6655_cast_fp16, var_6663_cast_fp16, var_6671_cast_fp16, var_6679_cast_fp16, var_6687_cast_fp16))[name = string("input_179_cast_fp16")]; string var_6696_pad_type_0 = const()[name = string("op_6696_pad_type_0"), val = string("valid")]; tensor var_6696_strides_0 = const()[name = string("op_6696_strides_0"), val = tensor([1, 1])]; tensor var_6696_pad_0 = const()[name = string("op_6696_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6696_dilations_0 = const()[name = string("op_6696_dilations_0"), val = tensor([1, 1])]; int32 var_6696_groups_0 = const()[name = string("op_6696_groups_0"), val = int32(1)]; tensor layers_21_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_21_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(559439552)))]; tensor layers_21_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_21_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(561536768)))]; tensor var_6696_cast_fp16 = conv(bias = layers_21_self_attn_out_proj_bias_to_fp16, dilations = var_6696_dilations_0, groups = var_6696_groups_0, pad = var_6696_pad_0, pad_type = var_6696_pad_type_0, strides = var_6696_strides_0, weight = layers_21_self_attn_out_proj_weight_to_fp16, x = input_179_cast_fp16)[name = string("op_6696_cast_fp16")]; tensor x_227_cast_fp16 = add(x = x_223_cast_fp16, y = var_6696_cast_fp16)[name = string("x_227_cast_fp16")]; tensor mu_87_axes_0 = const()[name = string("mu_87_axes_0"), val = tensor([1])]; bool mu_87_keep_dims_0 = const()[name = string("mu_87_keep_dims_0"), val = bool(true)]; tensor mu_87_cast_fp16 = reduce_mean(axes = mu_87_axes_0, keep_dims = mu_87_keep_dims_0, x = x_227_cast_fp16)[name = string("mu_87_cast_fp16")]; tensor var_6702_cast_fp16 = sub(x = x_227_cast_fp16, y = mu_87_cast_fp16)[name = string("op_6702_cast_fp16")]; fp16 var_6456_promoted_1_to_fp16 = const()[name = string("op_6456_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_6703_cast_fp16 = pow(x = var_6702_cast_fp16, y = var_6456_promoted_1_to_fp16)[name = string("op_6703_cast_fp16")]; tensor var_87_axes_0 = const()[name = string("var_87_axes_0"), val = tensor([1])]; bool var_87_keep_dims_0 = const()[name = string("var_87_keep_dims_0"), val = bool(true)]; tensor var_87_cast_fp16 = reduce_mean(axes = var_87_axes_0, keep_dims = var_87_keep_dims_0, x = var_6703_cast_fp16)[name = string("var_87_cast_fp16")]; fp16 var_6707_to_fp16 = const()[name = string("op_6707_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_6708_cast_fp16 = add(x = var_87_cast_fp16, y = var_6707_to_fp16)[name = string("op_6708_cast_fp16")]; fp32 var_6709_epsilon_0 = const()[name = string("op_6709_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_6709_cast_fp16 = rsqrt(epsilon = var_6709_epsilon_0, x = var_6708_cast_fp16)[name = string("op_6709_cast_fp16")]; tensor x_229_cast_fp16 = mul(x = var_6702_cast_fp16, y = var_6709_cast_fp16)[name = string("x_229_cast_fp16")]; tensor input_181_gamma_0_to_fp16 = const()[name = string("input_181_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(561538880)))]; tensor input_181_beta_0_to_fp16 = const()[name = string("input_181_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(561540992)))]; fp16 input_181_epsilon_0_to_fp16 = const()[name = string("input_181_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_181_cast_fp16 = batch_norm(beta = input_181_beta_0_to_fp16, epsilon = input_181_epsilon_0_to_fp16, gamma = input_181_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_229_cast_fp16)[name = string("input_181_cast_fp16")]; string x_231_pad_type_0 = const()[name = string("x_231_pad_type_0"), val = string("valid")]; tensor x_231_strides_0 = const()[name = string("x_231_strides_0"), val = tensor([1, 1])]; tensor x_231_pad_0 = const()[name = string("x_231_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_231_dilations_0 = const()[name = string("x_231_dilations_0"), val = tensor([1, 1])]; int32 x_231_groups_0 = const()[name = string("x_231_groups_0"), val = int32(1)]; tensor layers_21_fc1_weight_to_fp16 = const()[name = string("layers_21_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(561543104)))]; tensor layers_21_fc1_bias_to_fp16 = const()[name = string("layers_21_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(569931776)))]; tensor x_231_cast_fp16 = conv(bias = layers_21_fc1_bias_to_fp16, dilations = x_231_dilations_0, groups = x_231_groups_0, pad = x_231_pad_0, pad_type = x_231_pad_type_0, strides = x_231_strides_0, weight = layers_21_fc1_weight_to_fp16, x = input_181_cast_fp16)[name = string("x_231_cast_fp16")]; fp16 var_6724_to_fp16 = const()[name = string("op_6724_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_6725_cast_fp16 = mul(x = x_231_cast_fp16, y = var_6724_to_fp16)[name = string("op_6725_cast_fp16")]; tensor var_6726_cast_fp16 = mul(x = var_6725_cast_fp16, y = x_231_cast_fp16)[name = string("op_6726_cast_fp16")]; tensor var_6727_cast_fp16 = mul(x = var_6726_cast_fp16, y = x_231_cast_fp16)[name = string("op_6727_cast_fp16")]; tensor var_6728_cast_fp16 = add(x = x_231_cast_fp16, y = var_6727_cast_fp16)[name = string("op_6728_cast_fp16")]; fp16 var_6729_to_fp16 = const()[name = string("op_6729_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_49_cast_fp16 = mul(x = var_6728_cast_fp16, y = var_6729_to_fp16)[name = string("u_49_cast_fp16")]; fp16 var_6731_to_fp16 = const()[name = string("op_6731_to_fp16"), val = fp16(0x1p-1)]; tensor var_6732_cast_fp16 = mul(x = x_231_cast_fp16, y = var_6731_to_fp16)[name = string("op_6732_cast_fp16")]; tensor var_6733_cast_fp16 = tanh(x = u_49_cast_fp16)[name = string("op_6733_cast_fp16")]; fp16 var_6734_to_fp16 = const()[name = string("op_6734_to_fp16"), val = fp16(0x1p+0)]; tensor var_6735_cast_fp16 = add(x = var_6733_cast_fp16, y = var_6734_to_fp16)[name = string("op_6735_cast_fp16")]; tensor input_183_cast_fp16 = mul(x = var_6732_cast_fp16, y = var_6735_cast_fp16)[name = string("input_183_cast_fp16")]; string h_43_pad_type_0 = const()[name = string("h_43_pad_type_0"), val = string("valid")]; tensor h_43_strides_0 = const()[name = string("h_43_strides_0"), val = tensor([1, 1])]; tensor h_43_pad_0 = const()[name = string("h_43_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_43_dilations_0 = const()[name = string("h_43_dilations_0"), val = tensor([1, 1])]; int32 h_43_groups_0 = const()[name = string("h_43_groups_0"), val = int32(1)]; tensor layers_21_fc2_weight_to_fp16 = const()[name = string("layers_21_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(569940032)))]; tensor layers_21_fc2_bias_to_fp16 = const()[name = string("layers_21_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(578328704)))]; tensor h_43_cast_fp16 = conv(bias = layers_21_fc2_bias_to_fp16, dilations = h_43_dilations_0, groups = h_43_groups_0, pad = h_43_pad_0, pad_type = h_43_pad_type_0, strides = h_43_strides_0, weight = layers_21_fc2_weight_to_fp16, x = input_183_cast_fp16)[name = string("h_43_cast_fp16")]; tensor x_233_cast_fp16 = add(x = x_227_cast_fp16, y = h_43_cast_fp16)[name = string("x_233_cast_fp16")]; int32 var_6751 = const()[name = string("op_6751"), val = int32(1)]; tensor mu_89_axes_0 = const()[name = string("mu_89_axes_0"), val = tensor([1])]; bool mu_89_keep_dims_0 = const()[name = string("mu_89_keep_dims_0"), val = bool(true)]; tensor mu_89_cast_fp16 = reduce_mean(axes = mu_89_axes_0, keep_dims = mu_89_keep_dims_0, x = x_233_cast_fp16)[name = string("mu_89_cast_fp16")]; tensor var_6765_cast_fp16 = sub(x = x_233_cast_fp16, y = mu_89_cast_fp16)[name = string("op_6765_cast_fp16")]; fp16 var_6754_promoted_to_fp16 = const()[name = string("op_6754_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_6766_cast_fp16 = pow(x = var_6765_cast_fp16, y = var_6754_promoted_to_fp16)[name = string("op_6766_cast_fp16")]; tensor var_89_axes_0 = const()[name = string("var_89_axes_0"), val = tensor([1])]; bool var_89_keep_dims_0 = const()[name = string("var_89_keep_dims_0"), val = bool(true)]; tensor var_89_cast_fp16 = reduce_mean(axes = var_89_axes_0, keep_dims = var_89_keep_dims_0, x = var_6766_cast_fp16)[name = string("var_89_cast_fp16")]; fp16 var_6770_to_fp16 = const()[name = string("op_6770_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_6771_cast_fp16 = add(x = var_89_cast_fp16, y = var_6770_to_fp16)[name = string("op_6771_cast_fp16")]; fp32 var_6772_epsilon_0 = const()[name = string("op_6772_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_6772_cast_fp16 = rsqrt(epsilon = var_6772_epsilon_0, x = var_6771_cast_fp16)[name = string("op_6772_cast_fp16")]; tensor x_235_cast_fp16 = mul(x = var_6765_cast_fp16, y = var_6772_cast_fp16)[name = string("x_235_cast_fp16")]; tensor input_185_gamma_0_to_fp16 = const()[name = string("input_185_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(578330816)))]; tensor input_185_beta_0_to_fp16 = const()[name = string("input_185_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(578332928)))]; fp16 input_185_epsilon_0_to_fp16 = const()[name = string("input_185_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_185_cast_fp16 = batch_norm(beta = input_185_beta_0_to_fp16, epsilon = input_185_epsilon_0_to_fp16, gamma = input_185_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_235_cast_fp16)[name = string("input_185_cast_fp16")]; string var_6790_pad_type_0 = const()[name = string("op_6790_pad_type_0"), val = string("valid")]; tensor var_6790_strides_0 = const()[name = string("op_6790_strides_0"), val = tensor([1, 1])]; tensor var_6790_pad_0 = const()[name = string("op_6790_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6790_dilations_0 = const()[name = string("op_6790_dilations_0"), val = tensor([1, 1])]; int32 var_6790_groups_0 = const()[name = string("op_6790_groups_0"), val = int32(1)]; tensor var_6792_weight_0_to_fp16 = const()[name = string("op_6792_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(578335040)))]; tensor var_6792_bias_0_to_fp16 = const()[name = string("op_6792_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(580432256)))]; tensor var_6792_cast_fp16 = conv(bias = var_6792_bias_0_to_fp16, dilations = var_6790_dilations_0, groups = var_6790_groups_0, pad = var_6790_pad_0, pad_type = var_6790_pad_type_0, strides = var_6790_strides_0, weight = var_6792_weight_0_to_fp16, x = input_185_cast_fp16)[name = string("op_6792_cast_fp16")]; string var_6799_pad_type_0 = const()[name = string("op_6799_pad_type_0"), val = string("valid")]; tensor var_6799_strides_0 = const()[name = string("op_6799_strides_0"), val = tensor([1, 1])]; tensor var_6799_pad_0 = const()[name = string("op_6799_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6799_dilations_0 = const()[name = string("op_6799_dilations_0"), val = tensor([1, 1])]; int32 var_6799_groups_0 = const()[name = string("op_6799_groups_0"), val = int32(1)]; tensor layers_22_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_22_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(580434368)))]; tensor layers_22_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_22_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(582531584)))]; tensor var_6799_cast_fp16 = conv(bias = layers_22_self_attn_k_proj_bias_to_fp16, dilations = var_6799_dilations_0, groups = var_6799_groups_0, pad = var_6799_pad_0, pad_type = var_6799_pad_type_0, strides = var_6799_strides_0, weight = layers_22_self_attn_k_proj_weight_to_fp16, x = input_185_cast_fp16)[name = string("op_6799_cast_fp16")]; string var_6806_pad_type_0 = const()[name = string("op_6806_pad_type_0"), val = string("valid")]; tensor var_6806_strides_0 = const()[name = string("op_6806_strides_0"), val = tensor([1, 1])]; tensor var_6806_pad_0 = const()[name = string("op_6806_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6806_dilations_0 = const()[name = string("op_6806_dilations_0"), val = tensor([1, 1])]; int32 var_6806_groups_0 = const()[name = string("op_6806_groups_0"), val = int32(1)]; tensor layers_22_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_22_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(582533696)))]; tensor layers_22_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_22_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(584630912)))]; tensor var_6806_cast_fp16 = conv(bias = layers_22_self_attn_v_proj_bias_to_fp16, dilations = var_6806_dilations_0, groups = var_6806_groups_0, pad = var_6806_pad_0, pad_type = var_6806_pad_type_0, strides = var_6806_strides_0, weight = layers_22_self_attn_v_proj_weight_to_fp16, x = input_185_cast_fp16)[name = string("op_6806_cast_fp16")]; tensor tile_66 = const()[name = string("tile_66"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(584633024)))]; int32 var_6807_axis_0 = const()[name = string("op_6807_axis_0"), val = int32(1)]; tensor var_6807_cast_fp16_0, tensor var_6807_cast_fp16_1, tensor var_6807_cast_fp16_2, tensor var_6807_cast_fp16_3, tensor var_6807_cast_fp16_4, tensor var_6807_cast_fp16_5, tensor var_6807_cast_fp16_6, tensor var_6807_cast_fp16_7, tensor var_6807_cast_fp16_8, tensor var_6807_cast_fp16_9, tensor var_6807_cast_fp16_10, tensor var_6807_cast_fp16_11, tensor var_6807_cast_fp16_12, tensor var_6807_cast_fp16_13, tensor var_6807_cast_fp16_14, tensor var_6807_cast_fp16_15 = split(axis = var_6807_axis_0, split_sizes = tile_66, x = var_6792_cast_fp16)[name = string("op_6807_cast_fp16")]; tensor tile_67 = const()[name = string("tile_67"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(584633152)))]; int32 var_6824_axis_0 = const()[name = string("op_6824_axis_0"), val = int32(1)]; tensor var_6824_cast_fp16_0, tensor var_6824_cast_fp16_1, tensor var_6824_cast_fp16_2, tensor var_6824_cast_fp16_3, tensor var_6824_cast_fp16_4, tensor var_6824_cast_fp16_5, tensor var_6824_cast_fp16_6, tensor var_6824_cast_fp16_7, tensor var_6824_cast_fp16_8, tensor var_6824_cast_fp16_9, tensor var_6824_cast_fp16_10, tensor var_6824_cast_fp16_11, tensor var_6824_cast_fp16_12, tensor var_6824_cast_fp16_13, tensor var_6824_cast_fp16_14, tensor var_6824_cast_fp16_15 = split(axis = var_6824_axis_0, split_sizes = tile_67, x = var_6799_cast_fp16)[name = string("op_6824_cast_fp16")]; tensor tile_68 = const()[name = string("tile_68"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(584633280)))]; int32 var_6841_axis_0 = const()[name = string("op_6841_axis_0"), val = int32(1)]; tensor var_6841_cast_fp16_0, tensor var_6841_cast_fp16_1, tensor var_6841_cast_fp16_2, tensor var_6841_cast_fp16_3, tensor var_6841_cast_fp16_4, tensor var_6841_cast_fp16_5, tensor var_6841_cast_fp16_6, tensor var_6841_cast_fp16_7, tensor var_6841_cast_fp16_8, tensor var_6841_cast_fp16_9, tensor var_6841_cast_fp16_10, tensor var_6841_cast_fp16_11, tensor var_6841_cast_fp16_12, tensor var_6841_cast_fp16_13, tensor var_6841_cast_fp16_14, tensor var_6841_cast_fp16_15 = split(axis = var_6841_axis_0, split_sizes = tile_68, x = var_6806_cast_fp16)[name = string("op_6841_cast_fp16")]; tensor transpose_704_perm_0 = const()[name = string("transpose_704_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3524 = const()[name = string("concat_3524"), val = tensor([1, 104, 64])]; tensor transpose_704_cast_fp16 = transpose(perm = transpose_704_perm_0, x = var_6807_cast_fp16_0)[name = string("transpose_3167")]; tensor reshape_1056_cast_fp16 = reshape(shape = concat_3524, x = transpose_704_cast_fp16)[name = string("reshape_1056_cast_fp16")]; tensor transpose_705_perm_0 = const()[name = string("transpose_705_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3525 = const()[name = string("concat_3525"), val = tensor([1, 64, 104])]; tensor transpose_705_cast_fp16 = transpose(perm = transpose_705_perm_0, x = var_6824_cast_fp16_0)[name = string("transpose_3166")]; tensor reshape_1057_cast_fp16 = reshape(shape = concat_3525, x = transpose_705_cast_fp16)[name = string("reshape_1057_cast_fp16")]; bool matmul_352_transpose_x_0 = const()[name = string("matmul_352_transpose_x_0"), val = bool(false)]; bool matmul_352_transpose_y_0 = const()[name = string("matmul_352_transpose_y_0"), val = bool(false)]; tensor matmul_352_cast_fp16 = matmul(transpose_x = matmul_352_transpose_x_0, transpose_y = matmul_352_transpose_y_0, x = reshape_1056_cast_fp16, y = reshape_1057_cast_fp16)[name = string("matmul_352_cast_fp16")]; tensor concat_3529 = const()[name = string("concat_3529"), val = tensor([1, 1, 104, 104])]; tensor reshape_1058_cast_fp16 = reshape(shape = concat_3529, x = matmul_352_cast_fp16)[name = string("reshape_1058_cast_fp16")]; tensor transpose_3040_perm_0 = const()[name = string("transpose_3040_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3040 = transpose(perm = transpose_3040_perm_0, x = reshape_1058_cast_fp16)[name = string("transpose_3165")]; tensor w_1411_cast_fp16 = add(x = transpose_3040, y = transpose_2305)[name = string("w_1411_cast_fp16")]; tensor var_6863_cast_fp16 = softmax(axis = var_6751, x = w_1411_cast_fp16)[name = string("op_6863_cast_fp16")]; string var_6865_equation_0 = const()[name = string("op_6865_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6865_cast_fp16 = einsum(equation = var_6865_equation_0, values = (var_6841_cast_fp16_0, var_6863_cast_fp16))[name = string("op_6865_cast_fp16")]; tensor transpose_706_perm_0 = const()[name = string("transpose_706_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3534 = const()[name = string("concat_3534"), val = tensor([1, 104, 64])]; tensor transpose_706_cast_fp16 = transpose(perm = transpose_706_perm_0, x = var_6807_cast_fp16_1)[name = string("transpose_3164")]; tensor reshape_1059_cast_fp16 = reshape(shape = concat_3534, x = transpose_706_cast_fp16)[name = string("reshape_1059_cast_fp16")]; tensor transpose_707_perm_0 = const()[name = string("transpose_707_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3535 = const()[name = string("concat_3535"), val = tensor([1, 64, 104])]; tensor transpose_707_cast_fp16 = transpose(perm = transpose_707_perm_0, x = var_6824_cast_fp16_1)[name = string("transpose_3163")]; tensor reshape_1060_cast_fp16 = reshape(shape = concat_3535, x = transpose_707_cast_fp16)[name = string("reshape_1060_cast_fp16")]; bool matmul_353_transpose_x_0 = const()[name = string("matmul_353_transpose_x_0"), val = bool(false)]; bool matmul_353_transpose_y_0 = const()[name = string("matmul_353_transpose_y_0"), val = bool(false)]; tensor matmul_353_cast_fp16 = matmul(transpose_x = matmul_353_transpose_x_0, transpose_y = matmul_353_transpose_y_0, x = reshape_1059_cast_fp16, y = reshape_1060_cast_fp16)[name = string("matmul_353_cast_fp16")]; tensor concat_3539 = const()[name = string("concat_3539"), val = tensor([1, 1, 104, 104])]; tensor reshape_1061_cast_fp16 = reshape(shape = concat_3539, x = matmul_353_cast_fp16)[name = string("reshape_1061_cast_fp16")]; tensor transpose_3041_perm_0 = const()[name = string("transpose_3041_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3041 = transpose(perm = transpose_3041_perm_0, x = reshape_1061_cast_fp16)[name = string("transpose_3162")]; tensor w_1415_cast_fp16 = add(x = transpose_3041, y = transpose_2305)[name = string("w_1415_cast_fp16")]; tensor var_6871_cast_fp16 = softmax(axis = var_6751, x = w_1415_cast_fp16)[name = string("op_6871_cast_fp16")]; string var_6873_equation_0 = const()[name = string("op_6873_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6873_cast_fp16 = einsum(equation = var_6873_equation_0, values = (var_6841_cast_fp16_1, var_6871_cast_fp16))[name = string("op_6873_cast_fp16")]; tensor transpose_708_perm_0 = const()[name = string("transpose_708_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3544 = const()[name = string("concat_3544"), val = tensor([1, 104, 64])]; tensor transpose_708_cast_fp16 = transpose(perm = transpose_708_perm_0, x = var_6807_cast_fp16_2)[name = string("transpose_3161")]; tensor reshape_1062_cast_fp16 = reshape(shape = concat_3544, x = transpose_708_cast_fp16)[name = string("reshape_1062_cast_fp16")]; tensor transpose_709_perm_0 = const()[name = string("transpose_709_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3545 = const()[name = string("concat_3545"), val = tensor([1, 64, 104])]; tensor transpose_709_cast_fp16 = transpose(perm = transpose_709_perm_0, x = var_6824_cast_fp16_2)[name = string("transpose_3160")]; tensor reshape_1063_cast_fp16 = reshape(shape = concat_3545, x = transpose_709_cast_fp16)[name = string("reshape_1063_cast_fp16")]; bool matmul_354_transpose_x_0 = const()[name = string("matmul_354_transpose_x_0"), val = bool(false)]; bool matmul_354_transpose_y_0 = const()[name = string("matmul_354_transpose_y_0"), val = bool(false)]; tensor matmul_354_cast_fp16 = matmul(transpose_x = matmul_354_transpose_x_0, transpose_y = matmul_354_transpose_y_0, x = reshape_1062_cast_fp16, y = reshape_1063_cast_fp16)[name = string("matmul_354_cast_fp16")]; tensor concat_3549 = const()[name = string("concat_3549"), val = tensor([1, 1, 104, 104])]; tensor reshape_1064_cast_fp16 = reshape(shape = concat_3549, x = matmul_354_cast_fp16)[name = string("reshape_1064_cast_fp16")]; tensor transpose_3042_perm_0 = const()[name = string("transpose_3042_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3042 = transpose(perm = transpose_3042_perm_0, x = reshape_1064_cast_fp16)[name = string("transpose_3159")]; tensor w_1419_cast_fp16 = add(x = transpose_3042, y = transpose_2305)[name = string("w_1419_cast_fp16")]; tensor var_6879_cast_fp16 = softmax(axis = var_6751, x = w_1419_cast_fp16)[name = string("op_6879_cast_fp16")]; string var_6881_equation_0 = const()[name = string("op_6881_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6881_cast_fp16 = einsum(equation = var_6881_equation_0, values = (var_6841_cast_fp16_2, var_6879_cast_fp16))[name = string("op_6881_cast_fp16")]; tensor transpose_710_perm_0 = const()[name = string("transpose_710_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3554 = const()[name = string("concat_3554"), val = tensor([1, 104, 64])]; tensor transpose_710_cast_fp16 = transpose(perm = transpose_710_perm_0, x = var_6807_cast_fp16_3)[name = string("transpose_3158")]; tensor reshape_1065_cast_fp16 = reshape(shape = concat_3554, x = transpose_710_cast_fp16)[name = string("reshape_1065_cast_fp16")]; tensor transpose_711_perm_0 = const()[name = string("transpose_711_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3555 = const()[name = string("concat_3555"), val = tensor([1, 64, 104])]; tensor transpose_711_cast_fp16 = transpose(perm = transpose_711_perm_0, x = var_6824_cast_fp16_3)[name = string("transpose_3157")]; tensor reshape_1066_cast_fp16 = reshape(shape = concat_3555, x = transpose_711_cast_fp16)[name = string("reshape_1066_cast_fp16")]; bool matmul_355_transpose_x_0 = const()[name = string("matmul_355_transpose_x_0"), val = bool(false)]; bool matmul_355_transpose_y_0 = const()[name = string("matmul_355_transpose_y_0"), val = bool(false)]; tensor matmul_355_cast_fp16 = matmul(transpose_x = matmul_355_transpose_x_0, transpose_y = matmul_355_transpose_y_0, x = reshape_1065_cast_fp16, y = reshape_1066_cast_fp16)[name = string("matmul_355_cast_fp16")]; tensor concat_3559 = const()[name = string("concat_3559"), val = tensor([1, 1, 104, 104])]; tensor reshape_1067_cast_fp16 = reshape(shape = concat_3559, x = matmul_355_cast_fp16)[name = string("reshape_1067_cast_fp16")]; tensor transpose_3043_perm_0 = const()[name = string("transpose_3043_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3043 = transpose(perm = transpose_3043_perm_0, x = reshape_1067_cast_fp16)[name = string("transpose_3156")]; tensor w_1423_cast_fp16 = add(x = transpose_3043, y = transpose_2305)[name = string("w_1423_cast_fp16")]; tensor var_6887_cast_fp16 = softmax(axis = var_6751, x = w_1423_cast_fp16)[name = string("op_6887_cast_fp16")]; string var_6889_equation_0 = const()[name = string("op_6889_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6889_cast_fp16 = einsum(equation = var_6889_equation_0, values = (var_6841_cast_fp16_3, var_6887_cast_fp16))[name = string("op_6889_cast_fp16")]; tensor transpose_712_perm_0 = const()[name = string("transpose_712_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3564 = const()[name = string("concat_3564"), val = tensor([1, 104, 64])]; tensor transpose_712_cast_fp16 = transpose(perm = transpose_712_perm_0, x = var_6807_cast_fp16_4)[name = string("transpose_3155")]; tensor reshape_1068_cast_fp16 = reshape(shape = concat_3564, x = transpose_712_cast_fp16)[name = string("reshape_1068_cast_fp16")]; tensor transpose_713_perm_0 = const()[name = string("transpose_713_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3565 = const()[name = string("concat_3565"), val = tensor([1, 64, 104])]; tensor transpose_713_cast_fp16 = transpose(perm = transpose_713_perm_0, x = var_6824_cast_fp16_4)[name = string("transpose_3154")]; tensor reshape_1069_cast_fp16 = reshape(shape = concat_3565, x = transpose_713_cast_fp16)[name = string("reshape_1069_cast_fp16")]; bool matmul_356_transpose_x_0 = const()[name = string("matmul_356_transpose_x_0"), val = bool(false)]; bool matmul_356_transpose_y_0 = const()[name = string("matmul_356_transpose_y_0"), val = bool(false)]; tensor matmul_356_cast_fp16 = matmul(transpose_x = matmul_356_transpose_x_0, transpose_y = matmul_356_transpose_y_0, x = reshape_1068_cast_fp16, y = reshape_1069_cast_fp16)[name = string("matmul_356_cast_fp16")]; tensor concat_3569 = const()[name = string("concat_3569"), val = tensor([1, 1, 104, 104])]; tensor reshape_1070_cast_fp16 = reshape(shape = concat_3569, x = matmul_356_cast_fp16)[name = string("reshape_1070_cast_fp16")]; tensor transpose_3044_perm_0 = const()[name = string("transpose_3044_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3044 = transpose(perm = transpose_3044_perm_0, x = reshape_1070_cast_fp16)[name = string("transpose_3153")]; tensor w_1427_cast_fp16 = add(x = transpose_3044, y = transpose_2305)[name = string("w_1427_cast_fp16")]; tensor var_6895_cast_fp16 = softmax(axis = var_6751, x = w_1427_cast_fp16)[name = string("op_6895_cast_fp16")]; string var_6897_equation_0 = const()[name = string("op_6897_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6897_cast_fp16 = einsum(equation = var_6897_equation_0, values = (var_6841_cast_fp16_4, var_6895_cast_fp16))[name = string("op_6897_cast_fp16")]; tensor transpose_714_perm_0 = const()[name = string("transpose_714_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3574 = const()[name = string("concat_3574"), val = tensor([1, 104, 64])]; tensor transpose_714_cast_fp16 = transpose(perm = transpose_714_perm_0, x = var_6807_cast_fp16_5)[name = string("transpose_3152")]; tensor reshape_1071_cast_fp16 = reshape(shape = concat_3574, x = transpose_714_cast_fp16)[name = string("reshape_1071_cast_fp16")]; tensor transpose_715_perm_0 = const()[name = string("transpose_715_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3575 = const()[name = string("concat_3575"), val = tensor([1, 64, 104])]; tensor transpose_715_cast_fp16 = transpose(perm = transpose_715_perm_0, x = var_6824_cast_fp16_5)[name = string("transpose_3151")]; tensor reshape_1072_cast_fp16 = reshape(shape = concat_3575, x = transpose_715_cast_fp16)[name = string("reshape_1072_cast_fp16")]; bool matmul_357_transpose_x_0 = const()[name = string("matmul_357_transpose_x_0"), val = bool(false)]; bool matmul_357_transpose_y_0 = const()[name = string("matmul_357_transpose_y_0"), val = bool(false)]; tensor matmul_357_cast_fp16 = matmul(transpose_x = matmul_357_transpose_x_0, transpose_y = matmul_357_transpose_y_0, x = reshape_1071_cast_fp16, y = reshape_1072_cast_fp16)[name = string("matmul_357_cast_fp16")]; tensor concat_3579 = const()[name = string("concat_3579"), val = tensor([1, 1, 104, 104])]; tensor reshape_1073_cast_fp16 = reshape(shape = concat_3579, x = matmul_357_cast_fp16)[name = string("reshape_1073_cast_fp16")]; tensor transpose_3045_perm_0 = const()[name = string("transpose_3045_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3045 = transpose(perm = transpose_3045_perm_0, x = reshape_1073_cast_fp16)[name = string("transpose_3150")]; tensor w_1431_cast_fp16 = add(x = transpose_3045, y = transpose_2305)[name = string("w_1431_cast_fp16")]; tensor var_6903_cast_fp16 = softmax(axis = var_6751, x = w_1431_cast_fp16)[name = string("op_6903_cast_fp16")]; string var_6905_equation_0 = const()[name = string("op_6905_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6905_cast_fp16 = einsum(equation = var_6905_equation_0, values = (var_6841_cast_fp16_5, var_6903_cast_fp16))[name = string("op_6905_cast_fp16")]; tensor transpose_716_perm_0 = const()[name = string("transpose_716_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3584 = const()[name = string("concat_3584"), val = tensor([1, 104, 64])]; tensor transpose_716_cast_fp16 = transpose(perm = transpose_716_perm_0, x = var_6807_cast_fp16_6)[name = string("transpose_3149")]; tensor reshape_1074_cast_fp16 = reshape(shape = concat_3584, x = transpose_716_cast_fp16)[name = string("reshape_1074_cast_fp16")]; tensor transpose_717_perm_0 = const()[name = string("transpose_717_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3585 = const()[name = string("concat_3585"), val = tensor([1, 64, 104])]; tensor transpose_717_cast_fp16 = transpose(perm = transpose_717_perm_0, x = var_6824_cast_fp16_6)[name = string("transpose_3148")]; tensor reshape_1075_cast_fp16 = reshape(shape = concat_3585, x = transpose_717_cast_fp16)[name = string("reshape_1075_cast_fp16")]; bool matmul_358_transpose_x_0 = const()[name = string("matmul_358_transpose_x_0"), val = bool(false)]; bool matmul_358_transpose_y_0 = const()[name = string("matmul_358_transpose_y_0"), val = bool(false)]; tensor matmul_358_cast_fp16 = matmul(transpose_x = matmul_358_transpose_x_0, transpose_y = matmul_358_transpose_y_0, x = reshape_1074_cast_fp16, y = reshape_1075_cast_fp16)[name = string("matmul_358_cast_fp16")]; tensor concat_3589 = const()[name = string("concat_3589"), val = tensor([1, 1, 104, 104])]; tensor reshape_1076_cast_fp16 = reshape(shape = concat_3589, x = matmul_358_cast_fp16)[name = string("reshape_1076_cast_fp16")]; tensor transpose_3046_perm_0 = const()[name = string("transpose_3046_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3046 = transpose(perm = transpose_3046_perm_0, x = reshape_1076_cast_fp16)[name = string("transpose_3147")]; tensor w_1435_cast_fp16 = add(x = transpose_3046, y = transpose_2305)[name = string("w_1435_cast_fp16")]; tensor var_6911_cast_fp16 = softmax(axis = var_6751, x = w_1435_cast_fp16)[name = string("op_6911_cast_fp16")]; string var_6913_equation_0 = const()[name = string("op_6913_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6913_cast_fp16 = einsum(equation = var_6913_equation_0, values = (var_6841_cast_fp16_6, var_6911_cast_fp16))[name = string("op_6913_cast_fp16")]; tensor transpose_718_perm_0 = const()[name = string("transpose_718_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3594 = const()[name = string("concat_3594"), val = tensor([1, 104, 64])]; tensor transpose_718_cast_fp16 = transpose(perm = transpose_718_perm_0, x = var_6807_cast_fp16_7)[name = string("transpose_3146")]; tensor reshape_1077_cast_fp16 = reshape(shape = concat_3594, x = transpose_718_cast_fp16)[name = string("reshape_1077_cast_fp16")]; tensor transpose_719_perm_0 = const()[name = string("transpose_719_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3595 = const()[name = string("concat_3595"), val = tensor([1, 64, 104])]; tensor transpose_719_cast_fp16 = transpose(perm = transpose_719_perm_0, x = var_6824_cast_fp16_7)[name = string("transpose_3145")]; tensor reshape_1078_cast_fp16 = reshape(shape = concat_3595, x = transpose_719_cast_fp16)[name = string("reshape_1078_cast_fp16")]; bool matmul_359_transpose_x_0 = const()[name = string("matmul_359_transpose_x_0"), val = bool(false)]; bool matmul_359_transpose_y_0 = const()[name = string("matmul_359_transpose_y_0"), val = bool(false)]; tensor matmul_359_cast_fp16 = matmul(transpose_x = matmul_359_transpose_x_0, transpose_y = matmul_359_transpose_y_0, x = reshape_1077_cast_fp16, y = reshape_1078_cast_fp16)[name = string("matmul_359_cast_fp16")]; tensor concat_3599 = const()[name = string("concat_3599"), val = tensor([1, 1, 104, 104])]; tensor reshape_1079_cast_fp16 = reshape(shape = concat_3599, x = matmul_359_cast_fp16)[name = string("reshape_1079_cast_fp16")]; tensor transpose_3047_perm_0 = const()[name = string("transpose_3047_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3047 = transpose(perm = transpose_3047_perm_0, x = reshape_1079_cast_fp16)[name = string("transpose_3144")]; tensor w_1439_cast_fp16 = add(x = transpose_3047, y = transpose_2305)[name = string("w_1439_cast_fp16")]; tensor var_6919_cast_fp16 = softmax(axis = var_6751, x = w_1439_cast_fp16)[name = string("op_6919_cast_fp16")]; string var_6921_equation_0 = const()[name = string("op_6921_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6921_cast_fp16 = einsum(equation = var_6921_equation_0, values = (var_6841_cast_fp16_7, var_6919_cast_fp16))[name = string("op_6921_cast_fp16")]; tensor transpose_720_perm_0 = const()[name = string("transpose_720_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3604 = const()[name = string("concat_3604"), val = tensor([1, 104, 64])]; tensor transpose_720_cast_fp16 = transpose(perm = transpose_720_perm_0, x = var_6807_cast_fp16_8)[name = string("transpose_3143")]; tensor reshape_1080_cast_fp16 = reshape(shape = concat_3604, x = transpose_720_cast_fp16)[name = string("reshape_1080_cast_fp16")]; tensor transpose_721_perm_0 = const()[name = string("transpose_721_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3605 = const()[name = string("concat_3605"), val = tensor([1, 64, 104])]; tensor transpose_721_cast_fp16 = transpose(perm = transpose_721_perm_0, x = var_6824_cast_fp16_8)[name = string("transpose_3142")]; tensor reshape_1081_cast_fp16 = reshape(shape = concat_3605, x = transpose_721_cast_fp16)[name = string("reshape_1081_cast_fp16")]; bool matmul_360_transpose_x_0 = const()[name = string("matmul_360_transpose_x_0"), val = bool(false)]; bool matmul_360_transpose_y_0 = const()[name = string("matmul_360_transpose_y_0"), val = bool(false)]; tensor matmul_360_cast_fp16 = matmul(transpose_x = matmul_360_transpose_x_0, transpose_y = matmul_360_transpose_y_0, x = reshape_1080_cast_fp16, y = reshape_1081_cast_fp16)[name = string("matmul_360_cast_fp16")]; tensor concat_3609 = const()[name = string("concat_3609"), val = tensor([1, 1, 104, 104])]; tensor reshape_1082_cast_fp16 = reshape(shape = concat_3609, x = matmul_360_cast_fp16)[name = string("reshape_1082_cast_fp16")]; tensor transpose_3048_perm_0 = const()[name = string("transpose_3048_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3048 = transpose(perm = transpose_3048_perm_0, x = reshape_1082_cast_fp16)[name = string("transpose_3141")]; tensor w_1443_cast_fp16 = add(x = transpose_3048, y = transpose_2305)[name = string("w_1443_cast_fp16")]; tensor var_6927_cast_fp16 = softmax(axis = var_6751, x = w_1443_cast_fp16)[name = string("op_6927_cast_fp16")]; string var_6929_equation_0 = const()[name = string("op_6929_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6929_cast_fp16 = einsum(equation = var_6929_equation_0, values = (var_6841_cast_fp16_8, var_6927_cast_fp16))[name = string("op_6929_cast_fp16")]; tensor transpose_722_perm_0 = const()[name = string("transpose_722_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3614 = const()[name = string("concat_3614"), val = tensor([1, 104, 64])]; tensor transpose_722_cast_fp16 = transpose(perm = transpose_722_perm_0, x = var_6807_cast_fp16_9)[name = string("transpose_3140")]; tensor reshape_1083_cast_fp16 = reshape(shape = concat_3614, x = transpose_722_cast_fp16)[name = string("reshape_1083_cast_fp16")]; tensor transpose_723_perm_0 = const()[name = string("transpose_723_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3615 = const()[name = string("concat_3615"), val = tensor([1, 64, 104])]; tensor transpose_723_cast_fp16 = transpose(perm = transpose_723_perm_0, x = var_6824_cast_fp16_9)[name = string("transpose_3139")]; tensor reshape_1084_cast_fp16 = reshape(shape = concat_3615, x = transpose_723_cast_fp16)[name = string("reshape_1084_cast_fp16")]; bool matmul_361_transpose_x_0 = const()[name = string("matmul_361_transpose_x_0"), val = bool(false)]; bool matmul_361_transpose_y_0 = const()[name = string("matmul_361_transpose_y_0"), val = bool(false)]; tensor matmul_361_cast_fp16 = matmul(transpose_x = matmul_361_transpose_x_0, transpose_y = matmul_361_transpose_y_0, x = reshape_1083_cast_fp16, y = reshape_1084_cast_fp16)[name = string("matmul_361_cast_fp16")]; tensor concat_3619 = const()[name = string("concat_3619"), val = tensor([1, 1, 104, 104])]; tensor reshape_1085_cast_fp16 = reshape(shape = concat_3619, x = matmul_361_cast_fp16)[name = string("reshape_1085_cast_fp16")]; tensor transpose_3049_perm_0 = const()[name = string("transpose_3049_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3049 = transpose(perm = transpose_3049_perm_0, x = reshape_1085_cast_fp16)[name = string("transpose_3138")]; tensor w_1447_cast_fp16 = add(x = transpose_3049, y = transpose_2305)[name = string("w_1447_cast_fp16")]; tensor var_6935_cast_fp16 = softmax(axis = var_6751, x = w_1447_cast_fp16)[name = string("op_6935_cast_fp16")]; string var_6937_equation_0 = const()[name = string("op_6937_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6937_cast_fp16 = einsum(equation = var_6937_equation_0, values = (var_6841_cast_fp16_9, var_6935_cast_fp16))[name = string("op_6937_cast_fp16")]; tensor transpose_724_perm_0 = const()[name = string("transpose_724_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3624 = const()[name = string("concat_3624"), val = tensor([1, 104, 64])]; tensor transpose_724_cast_fp16 = transpose(perm = transpose_724_perm_0, x = var_6807_cast_fp16_10)[name = string("transpose_3137")]; tensor reshape_1086_cast_fp16 = reshape(shape = concat_3624, x = transpose_724_cast_fp16)[name = string("reshape_1086_cast_fp16")]; tensor transpose_725_perm_0 = const()[name = string("transpose_725_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3625 = const()[name = string("concat_3625"), val = tensor([1, 64, 104])]; tensor transpose_725_cast_fp16 = transpose(perm = transpose_725_perm_0, x = var_6824_cast_fp16_10)[name = string("transpose_3136")]; tensor reshape_1087_cast_fp16 = reshape(shape = concat_3625, x = transpose_725_cast_fp16)[name = string("reshape_1087_cast_fp16")]; bool matmul_362_transpose_x_0 = const()[name = string("matmul_362_transpose_x_0"), val = bool(false)]; bool matmul_362_transpose_y_0 = const()[name = string("matmul_362_transpose_y_0"), val = bool(false)]; tensor matmul_362_cast_fp16 = matmul(transpose_x = matmul_362_transpose_x_0, transpose_y = matmul_362_transpose_y_0, x = reshape_1086_cast_fp16, y = reshape_1087_cast_fp16)[name = string("matmul_362_cast_fp16")]; tensor concat_3629 = const()[name = string("concat_3629"), val = tensor([1, 1, 104, 104])]; tensor reshape_1088_cast_fp16 = reshape(shape = concat_3629, x = matmul_362_cast_fp16)[name = string("reshape_1088_cast_fp16")]; tensor transpose_3050_perm_0 = const()[name = string("transpose_3050_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3050 = transpose(perm = transpose_3050_perm_0, x = reshape_1088_cast_fp16)[name = string("transpose_3135")]; tensor w_1451_cast_fp16 = add(x = transpose_3050, y = transpose_2305)[name = string("w_1451_cast_fp16")]; tensor var_6943_cast_fp16 = softmax(axis = var_6751, x = w_1451_cast_fp16)[name = string("op_6943_cast_fp16")]; string var_6945_equation_0 = const()[name = string("op_6945_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6945_cast_fp16 = einsum(equation = var_6945_equation_0, values = (var_6841_cast_fp16_10, var_6943_cast_fp16))[name = string("op_6945_cast_fp16")]; tensor transpose_726_perm_0 = const()[name = string("transpose_726_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3634 = const()[name = string("concat_3634"), val = tensor([1, 104, 64])]; tensor transpose_726_cast_fp16 = transpose(perm = transpose_726_perm_0, x = var_6807_cast_fp16_11)[name = string("transpose_3134")]; tensor reshape_1089_cast_fp16 = reshape(shape = concat_3634, x = transpose_726_cast_fp16)[name = string("reshape_1089_cast_fp16")]; tensor transpose_727_perm_0 = const()[name = string("transpose_727_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3635 = const()[name = string("concat_3635"), val = tensor([1, 64, 104])]; tensor transpose_727_cast_fp16 = transpose(perm = transpose_727_perm_0, x = var_6824_cast_fp16_11)[name = string("transpose_3133")]; tensor reshape_1090_cast_fp16 = reshape(shape = concat_3635, x = transpose_727_cast_fp16)[name = string("reshape_1090_cast_fp16")]; bool matmul_363_transpose_x_0 = const()[name = string("matmul_363_transpose_x_0"), val = bool(false)]; bool matmul_363_transpose_y_0 = const()[name = string("matmul_363_transpose_y_0"), val = bool(false)]; tensor matmul_363_cast_fp16 = matmul(transpose_x = matmul_363_transpose_x_0, transpose_y = matmul_363_transpose_y_0, x = reshape_1089_cast_fp16, y = reshape_1090_cast_fp16)[name = string("matmul_363_cast_fp16")]; tensor concat_3639 = const()[name = string("concat_3639"), val = tensor([1, 1, 104, 104])]; tensor reshape_1091_cast_fp16 = reshape(shape = concat_3639, x = matmul_363_cast_fp16)[name = string("reshape_1091_cast_fp16")]; tensor transpose_3051_perm_0 = const()[name = string("transpose_3051_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3051 = transpose(perm = transpose_3051_perm_0, x = reshape_1091_cast_fp16)[name = string("transpose_3132")]; tensor w_1455_cast_fp16 = add(x = transpose_3051, y = transpose_2305)[name = string("w_1455_cast_fp16")]; tensor var_6951_cast_fp16 = softmax(axis = var_6751, x = w_1455_cast_fp16)[name = string("op_6951_cast_fp16")]; string var_6953_equation_0 = const()[name = string("op_6953_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6953_cast_fp16 = einsum(equation = var_6953_equation_0, values = (var_6841_cast_fp16_11, var_6951_cast_fp16))[name = string("op_6953_cast_fp16")]; tensor transpose_728_perm_0 = const()[name = string("transpose_728_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3644 = const()[name = string("concat_3644"), val = tensor([1, 104, 64])]; tensor transpose_728_cast_fp16 = transpose(perm = transpose_728_perm_0, x = var_6807_cast_fp16_12)[name = string("transpose_3131")]; tensor reshape_1092_cast_fp16 = reshape(shape = concat_3644, x = transpose_728_cast_fp16)[name = string("reshape_1092_cast_fp16")]; tensor transpose_729_perm_0 = const()[name = string("transpose_729_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3645 = const()[name = string("concat_3645"), val = tensor([1, 64, 104])]; tensor transpose_729_cast_fp16 = transpose(perm = transpose_729_perm_0, x = var_6824_cast_fp16_12)[name = string("transpose_3130")]; tensor reshape_1093_cast_fp16 = reshape(shape = concat_3645, x = transpose_729_cast_fp16)[name = string("reshape_1093_cast_fp16")]; bool matmul_364_transpose_x_0 = const()[name = string("matmul_364_transpose_x_0"), val = bool(false)]; bool matmul_364_transpose_y_0 = const()[name = string("matmul_364_transpose_y_0"), val = bool(false)]; tensor matmul_364_cast_fp16 = matmul(transpose_x = matmul_364_transpose_x_0, transpose_y = matmul_364_transpose_y_0, x = reshape_1092_cast_fp16, y = reshape_1093_cast_fp16)[name = string("matmul_364_cast_fp16")]; tensor concat_3649 = const()[name = string("concat_3649"), val = tensor([1, 1, 104, 104])]; tensor reshape_1094_cast_fp16 = reshape(shape = concat_3649, x = matmul_364_cast_fp16)[name = string("reshape_1094_cast_fp16")]; tensor transpose_3052_perm_0 = const()[name = string("transpose_3052_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3052 = transpose(perm = transpose_3052_perm_0, x = reshape_1094_cast_fp16)[name = string("transpose_3129")]; tensor w_1459_cast_fp16 = add(x = transpose_3052, y = transpose_2305)[name = string("w_1459_cast_fp16")]; tensor var_6959_cast_fp16 = softmax(axis = var_6751, x = w_1459_cast_fp16)[name = string("op_6959_cast_fp16")]; string var_6961_equation_0 = const()[name = string("op_6961_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6961_cast_fp16 = einsum(equation = var_6961_equation_0, values = (var_6841_cast_fp16_12, var_6959_cast_fp16))[name = string("op_6961_cast_fp16")]; tensor transpose_730_perm_0 = const()[name = string("transpose_730_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3654 = const()[name = string("concat_3654"), val = tensor([1, 104, 64])]; tensor transpose_730_cast_fp16 = transpose(perm = transpose_730_perm_0, x = var_6807_cast_fp16_13)[name = string("transpose_3128")]; tensor reshape_1095_cast_fp16 = reshape(shape = concat_3654, x = transpose_730_cast_fp16)[name = string("reshape_1095_cast_fp16")]; tensor transpose_731_perm_0 = const()[name = string("transpose_731_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3655 = const()[name = string("concat_3655"), val = tensor([1, 64, 104])]; tensor transpose_731_cast_fp16 = transpose(perm = transpose_731_perm_0, x = var_6824_cast_fp16_13)[name = string("transpose_3127")]; tensor reshape_1096_cast_fp16 = reshape(shape = concat_3655, x = transpose_731_cast_fp16)[name = string("reshape_1096_cast_fp16")]; bool matmul_365_transpose_x_0 = const()[name = string("matmul_365_transpose_x_0"), val = bool(false)]; bool matmul_365_transpose_y_0 = const()[name = string("matmul_365_transpose_y_0"), val = bool(false)]; tensor matmul_365_cast_fp16 = matmul(transpose_x = matmul_365_transpose_x_0, transpose_y = matmul_365_transpose_y_0, x = reshape_1095_cast_fp16, y = reshape_1096_cast_fp16)[name = string("matmul_365_cast_fp16")]; tensor concat_3659 = const()[name = string("concat_3659"), val = tensor([1, 1, 104, 104])]; tensor reshape_1097_cast_fp16 = reshape(shape = concat_3659, x = matmul_365_cast_fp16)[name = string("reshape_1097_cast_fp16")]; tensor transpose_3053_perm_0 = const()[name = string("transpose_3053_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3053 = transpose(perm = transpose_3053_perm_0, x = reshape_1097_cast_fp16)[name = string("transpose_3126")]; tensor w_1463_cast_fp16 = add(x = transpose_3053, y = transpose_2305)[name = string("w_1463_cast_fp16")]; tensor var_6967_cast_fp16 = softmax(axis = var_6751, x = w_1463_cast_fp16)[name = string("op_6967_cast_fp16")]; string var_6969_equation_0 = const()[name = string("op_6969_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6969_cast_fp16 = einsum(equation = var_6969_equation_0, values = (var_6841_cast_fp16_13, var_6967_cast_fp16))[name = string("op_6969_cast_fp16")]; tensor transpose_732_perm_0 = const()[name = string("transpose_732_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3664 = const()[name = string("concat_3664"), val = tensor([1, 104, 64])]; tensor transpose_732_cast_fp16 = transpose(perm = transpose_732_perm_0, x = var_6807_cast_fp16_14)[name = string("transpose_3125")]; tensor reshape_1098_cast_fp16 = reshape(shape = concat_3664, x = transpose_732_cast_fp16)[name = string("reshape_1098_cast_fp16")]; tensor transpose_733_perm_0 = const()[name = string("transpose_733_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3665 = const()[name = string("concat_3665"), val = tensor([1, 64, 104])]; tensor transpose_733_cast_fp16 = transpose(perm = transpose_733_perm_0, x = var_6824_cast_fp16_14)[name = string("transpose_3124")]; tensor reshape_1099_cast_fp16 = reshape(shape = concat_3665, x = transpose_733_cast_fp16)[name = string("reshape_1099_cast_fp16")]; bool matmul_366_transpose_x_0 = const()[name = string("matmul_366_transpose_x_0"), val = bool(false)]; bool matmul_366_transpose_y_0 = const()[name = string("matmul_366_transpose_y_0"), val = bool(false)]; tensor matmul_366_cast_fp16 = matmul(transpose_x = matmul_366_transpose_x_0, transpose_y = matmul_366_transpose_y_0, x = reshape_1098_cast_fp16, y = reshape_1099_cast_fp16)[name = string("matmul_366_cast_fp16")]; tensor concat_3669 = const()[name = string("concat_3669"), val = tensor([1, 1, 104, 104])]; tensor reshape_1100_cast_fp16 = reshape(shape = concat_3669, x = matmul_366_cast_fp16)[name = string("reshape_1100_cast_fp16")]; tensor transpose_3054_perm_0 = const()[name = string("transpose_3054_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3054 = transpose(perm = transpose_3054_perm_0, x = reshape_1100_cast_fp16)[name = string("transpose_3123")]; tensor w_1467_cast_fp16 = add(x = transpose_3054, y = transpose_2305)[name = string("w_1467_cast_fp16")]; tensor var_6975_cast_fp16 = softmax(axis = var_6751, x = w_1467_cast_fp16)[name = string("op_6975_cast_fp16")]; string var_6977_equation_0 = const()[name = string("op_6977_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6977_cast_fp16 = einsum(equation = var_6977_equation_0, values = (var_6841_cast_fp16_14, var_6975_cast_fp16))[name = string("op_6977_cast_fp16")]; tensor transpose_734_perm_0 = const()[name = string("transpose_734_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3674 = const()[name = string("concat_3674"), val = tensor([1, 104, 64])]; tensor transpose_734_cast_fp16 = transpose(perm = transpose_734_perm_0, x = var_6807_cast_fp16_15)[name = string("transpose_3122")]; tensor reshape_1101_cast_fp16 = reshape(shape = concat_3674, x = transpose_734_cast_fp16)[name = string("reshape_1101_cast_fp16")]; tensor transpose_735_perm_0 = const()[name = string("transpose_735_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3675 = const()[name = string("concat_3675"), val = tensor([1, 64, 104])]; tensor transpose_735_cast_fp16 = transpose(perm = transpose_735_perm_0, x = var_6824_cast_fp16_15)[name = string("transpose_3121")]; tensor reshape_1102_cast_fp16 = reshape(shape = concat_3675, x = transpose_735_cast_fp16)[name = string("reshape_1102_cast_fp16")]; bool matmul_367_transpose_x_0 = const()[name = string("matmul_367_transpose_x_0"), val = bool(false)]; bool matmul_367_transpose_y_0 = const()[name = string("matmul_367_transpose_y_0"), val = bool(false)]; tensor matmul_367_cast_fp16 = matmul(transpose_x = matmul_367_transpose_x_0, transpose_y = matmul_367_transpose_y_0, x = reshape_1101_cast_fp16, y = reshape_1102_cast_fp16)[name = string("matmul_367_cast_fp16")]; tensor concat_3679 = const()[name = string("concat_3679"), val = tensor([1, 1, 104, 104])]; tensor reshape_1103_cast_fp16 = reshape(shape = concat_3679, x = matmul_367_cast_fp16)[name = string("reshape_1103_cast_fp16")]; tensor transpose_3055_perm_0 = const()[name = string("transpose_3055_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3055 = transpose(perm = transpose_3055_perm_0, x = reshape_1103_cast_fp16)[name = string("transpose_3120")]; tensor w_1471_cast_fp16 = add(x = transpose_3055, y = transpose_2305)[name = string("w_1471_cast_fp16")]; tensor var_6983_cast_fp16 = softmax(axis = var_6751, x = w_1471_cast_fp16)[name = string("op_6983_cast_fp16")]; string var_6985_equation_0 = const()[name = string("op_6985_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_6985_cast_fp16 = einsum(equation = var_6985_equation_0, values = (var_6841_cast_fp16_15, var_6983_cast_fp16))[name = string("op_6985_cast_fp16")]; bool input_187_interleave_0 = const()[name = string("input_187_interleave_0"), val = bool(false)]; tensor input_187_cast_fp16 = concat(axis = var_6751, interleave = input_187_interleave_0, values = (var_6865_cast_fp16, var_6873_cast_fp16, var_6881_cast_fp16, var_6889_cast_fp16, var_6897_cast_fp16, var_6905_cast_fp16, var_6913_cast_fp16, var_6921_cast_fp16, var_6929_cast_fp16, var_6937_cast_fp16, var_6945_cast_fp16, var_6953_cast_fp16, var_6961_cast_fp16, var_6969_cast_fp16, var_6977_cast_fp16, var_6985_cast_fp16))[name = string("input_187_cast_fp16")]; string var_6994_pad_type_0 = const()[name = string("op_6994_pad_type_0"), val = string("valid")]; tensor var_6994_strides_0 = const()[name = string("op_6994_strides_0"), val = tensor([1, 1])]; tensor var_6994_pad_0 = const()[name = string("op_6994_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_6994_dilations_0 = const()[name = string("op_6994_dilations_0"), val = tensor([1, 1])]; int32 var_6994_groups_0 = const()[name = string("op_6994_groups_0"), val = int32(1)]; tensor layers_22_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_22_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(584633408)))]; tensor layers_22_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_22_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(586730624)))]; tensor var_6994_cast_fp16 = conv(bias = layers_22_self_attn_out_proj_bias_to_fp16, dilations = var_6994_dilations_0, groups = var_6994_groups_0, pad = var_6994_pad_0, pad_type = var_6994_pad_type_0, strides = var_6994_strides_0, weight = layers_22_self_attn_out_proj_weight_to_fp16, x = input_187_cast_fp16)[name = string("op_6994_cast_fp16")]; tensor x_237_cast_fp16 = add(x = x_233_cast_fp16, y = var_6994_cast_fp16)[name = string("x_237_cast_fp16")]; tensor mu_91_axes_0 = const()[name = string("mu_91_axes_0"), val = tensor([1])]; bool mu_91_keep_dims_0 = const()[name = string("mu_91_keep_dims_0"), val = bool(true)]; tensor mu_91_cast_fp16 = reduce_mean(axes = mu_91_axes_0, keep_dims = mu_91_keep_dims_0, x = x_237_cast_fp16)[name = string("mu_91_cast_fp16")]; tensor var_7000_cast_fp16 = sub(x = x_237_cast_fp16, y = mu_91_cast_fp16)[name = string("op_7000_cast_fp16")]; fp16 var_6754_promoted_1_to_fp16 = const()[name = string("op_6754_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_7001_cast_fp16 = pow(x = var_7000_cast_fp16, y = var_6754_promoted_1_to_fp16)[name = string("op_7001_cast_fp16")]; tensor var_91_axes_0 = const()[name = string("var_91_axes_0"), val = tensor([1])]; bool var_91_keep_dims_0 = const()[name = string("var_91_keep_dims_0"), val = bool(true)]; tensor var_91_cast_fp16_0 = reduce_mean(axes = var_91_axes_0, keep_dims = var_91_keep_dims_0, x = var_7001_cast_fp16)[name = string("var_91_cast_fp16")]; fp16 var_7005_to_fp16 = const()[name = string("op_7005_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_7006_cast_fp16 = add(x = var_91_cast_fp16_0, y = var_7005_to_fp16)[name = string("op_7006_cast_fp16")]; fp32 var_7007_epsilon_0 = const()[name = string("op_7007_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_7007_cast_fp16 = rsqrt(epsilon = var_7007_epsilon_0, x = var_7006_cast_fp16)[name = string("op_7007_cast_fp16")]; tensor x_239_cast_fp16 = mul(x = var_7000_cast_fp16, y = var_7007_cast_fp16)[name = string("x_239_cast_fp16")]; tensor input_189_gamma_0_to_fp16 = const()[name = string("input_189_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(586732736)))]; tensor input_189_beta_0_to_fp16 = const()[name = string("input_189_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(586734848)))]; fp16 input_189_epsilon_0_to_fp16 = const()[name = string("input_189_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_189_cast_fp16 = batch_norm(beta = input_189_beta_0_to_fp16, epsilon = input_189_epsilon_0_to_fp16, gamma = input_189_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_239_cast_fp16)[name = string("input_189_cast_fp16")]; string x_241_pad_type_0 = const()[name = string("x_241_pad_type_0"), val = string("valid")]; tensor x_241_strides_0 = const()[name = string("x_241_strides_0"), val = tensor([1, 1])]; tensor x_241_pad_0 = const()[name = string("x_241_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_241_dilations_0 = const()[name = string("x_241_dilations_0"), val = tensor([1, 1])]; int32 x_241_groups_0 = const()[name = string("x_241_groups_0"), val = int32(1)]; tensor layers_22_fc1_weight_to_fp16 = const()[name = string("layers_22_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(586736960)))]; tensor layers_22_fc1_bias_to_fp16 = const()[name = string("layers_22_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(595125632)))]; tensor x_241_cast_fp16 = conv(bias = layers_22_fc1_bias_to_fp16, dilations = x_241_dilations_0, groups = x_241_groups_0, pad = x_241_pad_0, pad_type = x_241_pad_type_0, strides = x_241_strides_0, weight = layers_22_fc1_weight_to_fp16, x = input_189_cast_fp16)[name = string("x_241_cast_fp16")]; fp16 var_7022_to_fp16 = const()[name = string("op_7022_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_7023_cast_fp16 = mul(x = x_241_cast_fp16, y = var_7022_to_fp16)[name = string("op_7023_cast_fp16")]; tensor var_7024_cast_fp16 = mul(x = var_7023_cast_fp16, y = x_241_cast_fp16)[name = string("op_7024_cast_fp16")]; tensor var_7025_cast_fp16 = mul(x = var_7024_cast_fp16, y = x_241_cast_fp16)[name = string("op_7025_cast_fp16")]; tensor var_7026_cast_fp16 = add(x = x_241_cast_fp16, y = var_7025_cast_fp16)[name = string("op_7026_cast_fp16")]; fp16 var_7027_to_fp16 = const()[name = string("op_7027_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_51_cast_fp16 = mul(x = var_7026_cast_fp16, y = var_7027_to_fp16)[name = string("u_51_cast_fp16")]; fp16 var_7029_to_fp16 = const()[name = string("op_7029_to_fp16"), val = fp16(0x1p-1)]; tensor var_7030_cast_fp16 = mul(x = x_241_cast_fp16, y = var_7029_to_fp16)[name = string("op_7030_cast_fp16")]; tensor var_7031_cast_fp16 = tanh(x = u_51_cast_fp16)[name = string("op_7031_cast_fp16")]; fp16 var_7032_to_fp16 = const()[name = string("op_7032_to_fp16"), val = fp16(0x1p+0)]; tensor var_7033_cast_fp16 = add(x = var_7031_cast_fp16, y = var_7032_to_fp16)[name = string("op_7033_cast_fp16")]; tensor input_191_cast_fp16 = mul(x = var_7030_cast_fp16, y = var_7033_cast_fp16)[name = string("input_191_cast_fp16")]; string h_45_pad_type_0 = const()[name = string("h_45_pad_type_0"), val = string("valid")]; tensor h_45_strides_0 = const()[name = string("h_45_strides_0"), val = tensor([1, 1])]; tensor h_45_pad_0 = const()[name = string("h_45_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_45_dilations_0 = const()[name = string("h_45_dilations_0"), val = tensor([1, 1])]; int32 h_45_groups_0 = const()[name = string("h_45_groups_0"), val = int32(1)]; tensor layers_22_fc2_weight_to_fp16 = const()[name = string("layers_22_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(595133888)))]; tensor layers_22_fc2_bias_to_fp16 = const()[name = string("layers_22_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(603522560)))]; tensor h_45_cast_fp16 = conv(bias = layers_22_fc2_bias_to_fp16, dilations = h_45_dilations_0, groups = h_45_groups_0, pad = h_45_pad_0, pad_type = h_45_pad_type_0, strides = h_45_strides_0, weight = layers_22_fc2_weight_to_fp16, x = input_191_cast_fp16)[name = string("h_45_cast_fp16")]; tensor x_243_cast_fp16 = add(x = x_237_cast_fp16, y = h_45_cast_fp16)[name = string("x_243_cast_fp16")]; int32 var_7049 = const()[name = string("op_7049"), val = int32(1)]; tensor mu_93_axes_0 = const()[name = string("mu_93_axes_0"), val = tensor([1])]; bool mu_93_keep_dims_0 = const()[name = string("mu_93_keep_dims_0"), val = bool(true)]; tensor mu_93_cast_fp16 = reduce_mean(axes = mu_93_axes_0, keep_dims = mu_93_keep_dims_0, x = x_243_cast_fp16)[name = string("mu_93_cast_fp16")]; tensor var_7063_cast_fp16 = sub(x = x_243_cast_fp16, y = mu_93_cast_fp16)[name = string("op_7063_cast_fp16")]; fp16 var_7052_promoted_to_fp16 = const()[name = string("op_7052_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_7064_cast_fp16 = pow(x = var_7063_cast_fp16, y = var_7052_promoted_to_fp16)[name = string("op_7064_cast_fp16")]; tensor var_93_axes_0 = const()[name = string("var_93_axes_0"), val = tensor([1])]; bool var_93_keep_dims_0 = const()[name = string("var_93_keep_dims_0"), val = bool(true)]; tensor var_93_cast_fp16 = reduce_mean(axes = var_93_axes_0, keep_dims = var_93_keep_dims_0, x = var_7064_cast_fp16)[name = string("var_93_cast_fp16")]; fp16 var_7068_to_fp16 = const()[name = string("op_7068_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_7069_cast_fp16 = add(x = var_93_cast_fp16, y = var_7068_to_fp16)[name = string("op_7069_cast_fp16")]; fp32 var_7070_epsilon_0 = const()[name = string("op_7070_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_7070_cast_fp16 = rsqrt(epsilon = var_7070_epsilon_0, x = var_7069_cast_fp16)[name = string("op_7070_cast_fp16")]; tensor x_245_cast_fp16 = mul(x = var_7063_cast_fp16, y = var_7070_cast_fp16)[name = string("x_245_cast_fp16")]; tensor input_193_gamma_0_to_fp16 = const()[name = string("input_193_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(603524672)))]; tensor input_193_beta_0_to_fp16 = const()[name = string("input_193_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(603526784)))]; fp16 input_193_epsilon_0_to_fp16 = const()[name = string("input_193_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_193_cast_fp16 = batch_norm(beta = input_193_beta_0_to_fp16, epsilon = input_193_epsilon_0_to_fp16, gamma = input_193_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_245_cast_fp16)[name = string("input_193_cast_fp16")]; string var_7088_pad_type_0 = const()[name = string("op_7088_pad_type_0"), val = string("valid")]; tensor var_7088_strides_0 = const()[name = string("op_7088_strides_0"), val = tensor([1, 1])]; tensor var_7088_pad_0 = const()[name = string("op_7088_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_7088_dilations_0 = const()[name = string("op_7088_dilations_0"), val = tensor([1, 1])]; int32 var_7088_groups_0 = const()[name = string("op_7088_groups_0"), val = int32(1)]; tensor var_7090_weight_0_to_fp16 = const()[name = string("op_7090_weight_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(603528896)))]; tensor var_7090_bias_0_to_fp16 = const()[name = string("op_7090_bias_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(605626112)))]; tensor var_7090_cast_fp16 = conv(bias = var_7090_bias_0_to_fp16, dilations = var_7088_dilations_0, groups = var_7088_groups_0, pad = var_7088_pad_0, pad_type = var_7088_pad_type_0, strides = var_7088_strides_0, weight = var_7090_weight_0_to_fp16, x = input_193_cast_fp16)[name = string("op_7090_cast_fp16")]; string var_7097_pad_type_0 = const()[name = string("op_7097_pad_type_0"), val = string("valid")]; tensor var_7097_strides_0 = const()[name = string("op_7097_strides_0"), val = tensor([1, 1])]; tensor var_7097_pad_0 = const()[name = string("op_7097_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_7097_dilations_0 = const()[name = string("op_7097_dilations_0"), val = tensor([1, 1])]; int32 var_7097_groups_0 = const()[name = string("op_7097_groups_0"), val = int32(1)]; tensor layers_23_self_attn_k_proj_weight_to_fp16 = const()[name = string("layers_23_self_attn_k_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(605628224)))]; tensor layers_23_self_attn_k_proj_bias_to_fp16 = const()[name = string("layers_23_self_attn_k_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(607725440)))]; tensor var_7097_cast_fp16 = conv(bias = layers_23_self_attn_k_proj_bias_to_fp16, dilations = var_7097_dilations_0, groups = var_7097_groups_0, pad = var_7097_pad_0, pad_type = var_7097_pad_type_0, strides = var_7097_strides_0, weight = layers_23_self_attn_k_proj_weight_to_fp16, x = input_193_cast_fp16)[name = string("op_7097_cast_fp16")]; string var_7104_pad_type_0 = const()[name = string("op_7104_pad_type_0"), val = string("valid")]; tensor var_7104_strides_0 = const()[name = string("op_7104_strides_0"), val = tensor([1, 1])]; tensor var_7104_pad_0 = const()[name = string("op_7104_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_7104_dilations_0 = const()[name = string("op_7104_dilations_0"), val = tensor([1, 1])]; int32 var_7104_groups_0 = const()[name = string("op_7104_groups_0"), val = int32(1)]; tensor layers_23_self_attn_v_proj_weight_to_fp16 = const()[name = string("layers_23_self_attn_v_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(607727552)))]; tensor layers_23_self_attn_v_proj_bias_to_fp16 = const()[name = string("layers_23_self_attn_v_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(609824768)))]; tensor var_7104_cast_fp16 = conv(bias = layers_23_self_attn_v_proj_bias_to_fp16, dilations = var_7104_dilations_0, groups = var_7104_groups_0, pad = var_7104_pad_0, pad_type = var_7104_pad_type_0, strides = var_7104_strides_0, weight = layers_23_self_attn_v_proj_weight_to_fp16, x = input_193_cast_fp16)[name = string("op_7104_cast_fp16")]; tensor tile_69 = const()[name = string("tile_69"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(609826880)))]; int32 var_7105_axis_0 = const()[name = string("op_7105_axis_0"), val = int32(1)]; tensor var_7105_cast_fp16_0, tensor var_7105_cast_fp16_1, tensor var_7105_cast_fp16_2, tensor var_7105_cast_fp16_3, tensor var_7105_cast_fp16_4, tensor var_7105_cast_fp16_5, tensor var_7105_cast_fp16_6, tensor var_7105_cast_fp16_7, tensor var_7105_cast_fp16_8, tensor var_7105_cast_fp16_9, tensor var_7105_cast_fp16_10, tensor var_7105_cast_fp16_11, tensor var_7105_cast_fp16_12, tensor var_7105_cast_fp16_13, tensor var_7105_cast_fp16_14, tensor var_7105_cast_fp16_15 = split(axis = var_7105_axis_0, split_sizes = tile_69, x = var_7090_cast_fp16)[name = string("op_7105_cast_fp16")]; tensor tile_70 = const()[name = string("tile_70"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(609827008)))]; int32 var_7122_axis_0 = const()[name = string("op_7122_axis_0"), val = int32(1)]; tensor var_7122_cast_fp16_0, tensor var_7122_cast_fp16_1, tensor var_7122_cast_fp16_2, tensor var_7122_cast_fp16_3, tensor var_7122_cast_fp16_4, tensor var_7122_cast_fp16_5, tensor var_7122_cast_fp16_6, tensor var_7122_cast_fp16_7, tensor var_7122_cast_fp16_8, tensor var_7122_cast_fp16_9, tensor var_7122_cast_fp16_10, tensor var_7122_cast_fp16_11, tensor var_7122_cast_fp16_12, tensor var_7122_cast_fp16_13, tensor var_7122_cast_fp16_14, tensor var_7122_cast_fp16_15 = split(axis = var_7122_axis_0, split_sizes = tile_70, x = var_7097_cast_fp16)[name = string("op_7122_cast_fp16")]; tensor tile_71 = const()[name = string("tile_71"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(609827136)))]; int32 var_7139_axis_0 = const()[name = string("op_7139_axis_0"), val = int32(1)]; tensor var_7139_cast_fp16_0, tensor var_7139_cast_fp16_1, tensor var_7139_cast_fp16_2, tensor var_7139_cast_fp16_3, tensor var_7139_cast_fp16_4, tensor var_7139_cast_fp16_5, tensor var_7139_cast_fp16_6, tensor var_7139_cast_fp16_7, tensor var_7139_cast_fp16_8, tensor var_7139_cast_fp16_9, tensor var_7139_cast_fp16_10, tensor var_7139_cast_fp16_11, tensor var_7139_cast_fp16_12, tensor var_7139_cast_fp16_13, tensor var_7139_cast_fp16_14, tensor var_7139_cast_fp16_15 = split(axis = var_7139_axis_0, split_sizes = tile_71, x = var_7104_cast_fp16)[name = string("op_7139_cast_fp16")]; tensor transpose_736_perm_0 = const()[name = string("transpose_736_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3684 = const()[name = string("concat_3684"), val = tensor([1, 104, 64])]; tensor transpose_736_cast_fp16 = transpose(perm = transpose_736_perm_0, x = var_7105_cast_fp16_0)[name = string("transpose_3119")]; tensor reshape_1104_cast_fp16 = reshape(shape = concat_3684, x = transpose_736_cast_fp16)[name = string("reshape_1104_cast_fp16")]; tensor transpose_737_perm_0 = const()[name = string("transpose_737_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3685 = const()[name = string("concat_3685"), val = tensor([1, 64, 104])]; tensor transpose_737_cast_fp16 = transpose(perm = transpose_737_perm_0, x = var_7122_cast_fp16_0)[name = string("transpose_3118")]; tensor reshape_1105_cast_fp16 = reshape(shape = concat_3685, x = transpose_737_cast_fp16)[name = string("reshape_1105_cast_fp16")]; bool matmul_368_transpose_x_0 = const()[name = string("matmul_368_transpose_x_0"), val = bool(false)]; bool matmul_368_transpose_y_0 = const()[name = string("matmul_368_transpose_y_0"), val = bool(false)]; tensor matmul_368_cast_fp16 = matmul(transpose_x = matmul_368_transpose_x_0, transpose_y = matmul_368_transpose_y_0, x = reshape_1104_cast_fp16, y = reshape_1105_cast_fp16)[name = string("matmul_368_cast_fp16")]; tensor concat_3689 = const()[name = string("concat_3689"), val = tensor([1, 1, 104, 104])]; tensor reshape_1106_cast_fp16 = reshape(shape = concat_3689, x = matmul_368_cast_fp16)[name = string("reshape_1106_cast_fp16")]; tensor transpose_3056_perm_0 = const()[name = string("transpose_3056_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3056 = transpose(perm = transpose_3056_perm_0, x = reshape_1106_cast_fp16)[name = string("transpose_3117")]; tensor w_1475_cast_fp16 = add(x = transpose_3056, y = transpose_2305)[name = string("w_1475_cast_fp16")]; tensor var_7161_cast_fp16 = softmax(axis = var_7049, x = w_1475_cast_fp16)[name = string("op_7161_cast_fp16")]; string var_7163_equation_0 = const()[name = string("op_7163_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7163_cast_fp16 = einsum(equation = var_7163_equation_0, values = (var_7139_cast_fp16_0, var_7161_cast_fp16))[name = string("op_7163_cast_fp16")]; tensor transpose_738_perm_0 = const()[name = string("transpose_738_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3694 = const()[name = string("concat_3694"), val = tensor([1, 104, 64])]; tensor transpose_738_cast_fp16 = transpose(perm = transpose_738_perm_0, x = var_7105_cast_fp16_1)[name = string("transpose_3116")]; tensor reshape_1107_cast_fp16 = reshape(shape = concat_3694, x = transpose_738_cast_fp16)[name = string("reshape_1107_cast_fp16")]; tensor transpose_739_perm_0 = const()[name = string("transpose_739_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3695 = const()[name = string("concat_3695"), val = tensor([1, 64, 104])]; tensor transpose_739_cast_fp16 = transpose(perm = transpose_739_perm_0, x = var_7122_cast_fp16_1)[name = string("transpose_3115")]; tensor reshape_1108_cast_fp16 = reshape(shape = concat_3695, x = transpose_739_cast_fp16)[name = string("reshape_1108_cast_fp16")]; bool matmul_369_transpose_x_0 = const()[name = string("matmul_369_transpose_x_0"), val = bool(false)]; bool matmul_369_transpose_y_0 = const()[name = string("matmul_369_transpose_y_0"), val = bool(false)]; tensor matmul_369_cast_fp16 = matmul(transpose_x = matmul_369_transpose_x_0, transpose_y = matmul_369_transpose_y_0, x = reshape_1107_cast_fp16, y = reshape_1108_cast_fp16)[name = string("matmul_369_cast_fp16")]; tensor concat_3699 = const()[name = string("concat_3699"), val = tensor([1, 1, 104, 104])]; tensor reshape_1109_cast_fp16 = reshape(shape = concat_3699, x = matmul_369_cast_fp16)[name = string("reshape_1109_cast_fp16")]; tensor transpose_3057_perm_0 = const()[name = string("transpose_3057_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3057 = transpose(perm = transpose_3057_perm_0, x = reshape_1109_cast_fp16)[name = string("transpose_3114")]; tensor w_1479_cast_fp16 = add(x = transpose_3057, y = transpose_2305)[name = string("w_1479_cast_fp16")]; tensor var_7169_cast_fp16 = softmax(axis = var_7049, x = w_1479_cast_fp16)[name = string("op_7169_cast_fp16")]; string var_7171_equation_0 = const()[name = string("op_7171_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7171_cast_fp16 = einsum(equation = var_7171_equation_0, values = (var_7139_cast_fp16_1, var_7169_cast_fp16))[name = string("op_7171_cast_fp16")]; tensor transpose_740_perm_0 = const()[name = string("transpose_740_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3704 = const()[name = string("concat_3704"), val = tensor([1, 104, 64])]; tensor transpose_740_cast_fp16 = transpose(perm = transpose_740_perm_0, x = var_7105_cast_fp16_2)[name = string("transpose_3113")]; tensor reshape_1110_cast_fp16 = reshape(shape = concat_3704, x = transpose_740_cast_fp16)[name = string("reshape_1110_cast_fp16")]; tensor transpose_741_perm_0 = const()[name = string("transpose_741_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3705 = const()[name = string("concat_3705"), val = tensor([1, 64, 104])]; tensor transpose_741_cast_fp16 = transpose(perm = transpose_741_perm_0, x = var_7122_cast_fp16_2)[name = string("transpose_3112")]; tensor reshape_1111_cast_fp16 = reshape(shape = concat_3705, x = transpose_741_cast_fp16)[name = string("reshape_1111_cast_fp16")]; bool matmul_370_transpose_x_0 = const()[name = string("matmul_370_transpose_x_0"), val = bool(false)]; bool matmul_370_transpose_y_0 = const()[name = string("matmul_370_transpose_y_0"), val = bool(false)]; tensor matmul_370_cast_fp16 = matmul(transpose_x = matmul_370_transpose_x_0, transpose_y = matmul_370_transpose_y_0, x = reshape_1110_cast_fp16, y = reshape_1111_cast_fp16)[name = string("matmul_370_cast_fp16")]; tensor concat_3709 = const()[name = string("concat_3709"), val = tensor([1, 1, 104, 104])]; tensor reshape_1112_cast_fp16 = reshape(shape = concat_3709, x = matmul_370_cast_fp16)[name = string("reshape_1112_cast_fp16")]; tensor transpose_3058_perm_0 = const()[name = string("transpose_3058_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3058 = transpose(perm = transpose_3058_perm_0, x = reshape_1112_cast_fp16)[name = string("transpose_3111")]; tensor w_1483_cast_fp16 = add(x = transpose_3058, y = transpose_2305)[name = string("w_1483_cast_fp16")]; tensor var_7177_cast_fp16 = softmax(axis = var_7049, x = w_1483_cast_fp16)[name = string("op_7177_cast_fp16")]; string var_7179_equation_0 = const()[name = string("op_7179_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7179_cast_fp16 = einsum(equation = var_7179_equation_0, values = (var_7139_cast_fp16_2, var_7177_cast_fp16))[name = string("op_7179_cast_fp16")]; tensor transpose_742_perm_0 = const()[name = string("transpose_742_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3714 = const()[name = string("concat_3714"), val = tensor([1, 104, 64])]; tensor transpose_742_cast_fp16 = transpose(perm = transpose_742_perm_0, x = var_7105_cast_fp16_3)[name = string("transpose_3110")]; tensor reshape_1113_cast_fp16 = reshape(shape = concat_3714, x = transpose_742_cast_fp16)[name = string("reshape_1113_cast_fp16")]; tensor transpose_743_perm_0 = const()[name = string("transpose_743_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3715 = const()[name = string("concat_3715"), val = tensor([1, 64, 104])]; tensor transpose_743_cast_fp16 = transpose(perm = transpose_743_perm_0, x = var_7122_cast_fp16_3)[name = string("transpose_3109")]; tensor reshape_1114_cast_fp16 = reshape(shape = concat_3715, x = transpose_743_cast_fp16)[name = string("reshape_1114_cast_fp16")]; bool matmul_371_transpose_x_0 = const()[name = string("matmul_371_transpose_x_0"), val = bool(false)]; bool matmul_371_transpose_y_0 = const()[name = string("matmul_371_transpose_y_0"), val = bool(false)]; tensor matmul_371_cast_fp16 = matmul(transpose_x = matmul_371_transpose_x_0, transpose_y = matmul_371_transpose_y_0, x = reshape_1113_cast_fp16, y = reshape_1114_cast_fp16)[name = string("matmul_371_cast_fp16")]; tensor concat_3719 = const()[name = string("concat_3719"), val = tensor([1, 1, 104, 104])]; tensor reshape_1115_cast_fp16 = reshape(shape = concat_3719, x = matmul_371_cast_fp16)[name = string("reshape_1115_cast_fp16")]; tensor transpose_3059_perm_0 = const()[name = string("transpose_3059_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3059 = transpose(perm = transpose_3059_perm_0, x = reshape_1115_cast_fp16)[name = string("transpose_3108")]; tensor w_1487_cast_fp16 = add(x = transpose_3059, y = transpose_2305)[name = string("w_1487_cast_fp16")]; tensor var_7185_cast_fp16 = softmax(axis = var_7049, x = w_1487_cast_fp16)[name = string("op_7185_cast_fp16")]; string var_7187_equation_0 = const()[name = string("op_7187_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7187_cast_fp16 = einsum(equation = var_7187_equation_0, values = (var_7139_cast_fp16_3, var_7185_cast_fp16))[name = string("op_7187_cast_fp16")]; tensor transpose_744_perm_0 = const()[name = string("transpose_744_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3724 = const()[name = string("concat_3724"), val = tensor([1, 104, 64])]; tensor transpose_744_cast_fp16 = transpose(perm = transpose_744_perm_0, x = var_7105_cast_fp16_4)[name = string("transpose_3107")]; tensor reshape_1116_cast_fp16 = reshape(shape = concat_3724, x = transpose_744_cast_fp16)[name = string("reshape_1116_cast_fp16")]; tensor transpose_745_perm_0 = const()[name = string("transpose_745_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3725 = const()[name = string("concat_3725"), val = tensor([1, 64, 104])]; tensor transpose_745_cast_fp16 = transpose(perm = transpose_745_perm_0, x = var_7122_cast_fp16_4)[name = string("transpose_3106")]; tensor reshape_1117_cast_fp16 = reshape(shape = concat_3725, x = transpose_745_cast_fp16)[name = string("reshape_1117_cast_fp16")]; bool matmul_372_transpose_x_0 = const()[name = string("matmul_372_transpose_x_0"), val = bool(false)]; bool matmul_372_transpose_y_0 = const()[name = string("matmul_372_transpose_y_0"), val = bool(false)]; tensor matmul_372_cast_fp16 = matmul(transpose_x = matmul_372_transpose_x_0, transpose_y = matmul_372_transpose_y_0, x = reshape_1116_cast_fp16, y = reshape_1117_cast_fp16)[name = string("matmul_372_cast_fp16")]; tensor concat_3729 = const()[name = string("concat_3729"), val = tensor([1, 1, 104, 104])]; tensor reshape_1118_cast_fp16 = reshape(shape = concat_3729, x = matmul_372_cast_fp16)[name = string("reshape_1118_cast_fp16")]; tensor transpose_3060_perm_0 = const()[name = string("transpose_3060_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3060 = transpose(perm = transpose_3060_perm_0, x = reshape_1118_cast_fp16)[name = string("transpose_3105")]; tensor w_1491_cast_fp16 = add(x = transpose_3060, y = transpose_2305)[name = string("w_1491_cast_fp16")]; tensor var_7193_cast_fp16 = softmax(axis = var_7049, x = w_1491_cast_fp16)[name = string("op_7193_cast_fp16")]; string var_7195_equation_0 = const()[name = string("op_7195_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7195_cast_fp16 = einsum(equation = var_7195_equation_0, values = (var_7139_cast_fp16_4, var_7193_cast_fp16))[name = string("op_7195_cast_fp16")]; tensor transpose_746_perm_0 = const()[name = string("transpose_746_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3734 = const()[name = string("concat_3734"), val = tensor([1, 104, 64])]; tensor transpose_746_cast_fp16 = transpose(perm = transpose_746_perm_0, x = var_7105_cast_fp16_5)[name = string("transpose_3104")]; tensor reshape_1119_cast_fp16 = reshape(shape = concat_3734, x = transpose_746_cast_fp16)[name = string("reshape_1119_cast_fp16")]; tensor transpose_747_perm_0 = const()[name = string("transpose_747_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3735 = const()[name = string("concat_3735"), val = tensor([1, 64, 104])]; tensor transpose_747_cast_fp16 = transpose(perm = transpose_747_perm_0, x = var_7122_cast_fp16_5)[name = string("transpose_3103")]; tensor reshape_1120_cast_fp16 = reshape(shape = concat_3735, x = transpose_747_cast_fp16)[name = string("reshape_1120_cast_fp16")]; bool matmul_373_transpose_x_0 = const()[name = string("matmul_373_transpose_x_0"), val = bool(false)]; bool matmul_373_transpose_y_0 = const()[name = string("matmul_373_transpose_y_0"), val = bool(false)]; tensor matmul_373_cast_fp16 = matmul(transpose_x = matmul_373_transpose_x_0, transpose_y = matmul_373_transpose_y_0, x = reshape_1119_cast_fp16, y = reshape_1120_cast_fp16)[name = string("matmul_373_cast_fp16")]; tensor concat_3739 = const()[name = string("concat_3739"), val = tensor([1, 1, 104, 104])]; tensor reshape_1121_cast_fp16 = reshape(shape = concat_3739, x = matmul_373_cast_fp16)[name = string("reshape_1121_cast_fp16")]; tensor transpose_3061_perm_0 = const()[name = string("transpose_3061_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3061 = transpose(perm = transpose_3061_perm_0, x = reshape_1121_cast_fp16)[name = string("transpose_3102")]; tensor w_1495_cast_fp16 = add(x = transpose_3061, y = transpose_2305)[name = string("w_1495_cast_fp16")]; tensor var_7201_cast_fp16 = softmax(axis = var_7049, x = w_1495_cast_fp16)[name = string("op_7201_cast_fp16")]; string var_7203_equation_0 = const()[name = string("op_7203_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7203_cast_fp16 = einsum(equation = var_7203_equation_0, values = (var_7139_cast_fp16_5, var_7201_cast_fp16))[name = string("op_7203_cast_fp16")]; tensor transpose_748_perm_0 = const()[name = string("transpose_748_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3744 = const()[name = string("concat_3744"), val = tensor([1, 104, 64])]; tensor transpose_748_cast_fp16 = transpose(perm = transpose_748_perm_0, x = var_7105_cast_fp16_6)[name = string("transpose_3101")]; tensor reshape_1122_cast_fp16 = reshape(shape = concat_3744, x = transpose_748_cast_fp16)[name = string("reshape_1122_cast_fp16")]; tensor transpose_749_perm_0 = const()[name = string("transpose_749_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3745 = const()[name = string("concat_3745"), val = tensor([1, 64, 104])]; tensor transpose_749_cast_fp16 = transpose(perm = transpose_749_perm_0, x = var_7122_cast_fp16_6)[name = string("transpose_3100")]; tensor reshape_1123_cast_fp16 = reshape(shape = concat_3745, x = transpose_749_cast_fp16)[name = string("reshape_1123_cast_fp16")]; bool matmul_374_transpose_x_0 = const()[name = string("matmul_374_transpose_x_0"), val = bool(false)]; bool matmul_374_transpose_y_0 = const()[name = string("matmul_374_transpose_y_0"), val = bool(false)]; tensor matmul_374_cast_fp16 = matmul(transpose_x = matmul_374_transpose_x_0, transpose_y = matmul_374_transpose_y_0, x = reshape_1122_cast_fp16, y = reshape_1123_cast_fp16)[name = string("matmul_374_cast_fp16")]; tensor concat_3749 = const()[name = string("concat_3749"), val = tensor([1, 1, 104, 104])]; tensor reshape_1124_cast_fp16 = reshape(shape = concat_3749, x = matmul_374_cast_fp16)[name = string("reshape_1124_cast_fp16")]; tensor transpose_3062_perm_0 = const()[name = string("transpose_3062_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3062 = transpose(perm = transpose_3062_perm_0, x = reshape_1124_cast_fp16)[name = string("transpose_3099")]; tensor w_1499_cast_fp16 = add(x = transpose_3062, y = transpose_2305)[name = string("w_1499_cast_fp16")]; tensor var_7209_cast_fp16 = softmax(axis = var_7049, x = w_1499_cast_fp16)[name = string("op_7209_cast_fp16")]; string var_7211_equation_0 = const()[name = string("op_7211_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7211_cast_fp16 = einsum(equation = var_7211_equation_0, values = (var_7139_cast_fp16_6, var_7209_cast_fp16))[name = string("op_7211_cast_fp16")]; tensor transpose_750_perm_0 = const()[name = string("transpose_750_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3754 = const()[name = string("concat_3754"), val = tensor([1, 104, 64])]; tensor transpose_750_cast_fp16 = transpose(perm = transpose_750_perm_0, x = var_7105_cast_fp16_7)[name = string("transpose_3098")]; tensor reshape_1125_cast_fp16 = reshape(shape = concat_3754, x = transpose_750_cast_fp16)[name = string("reshape_1125_cast_fp16")]; tensor transpose_751_perm_0 = const()[name = string("transpose_751_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3755 = const()[name = string("concat_3755"), val = tensor([1, 64, 104])]; tensor transpose_751_cast_fp16 = transpose(perm = transpose_751_perm_0, x = var_7122_cast_fp16_7)[name = string("transpose_3097")]; tensor reshape_1126_cast_fp16 = reshape(shape = concat_3755, x = transpose_751_cast_fp16)[name = string("reshape_1126_cast_fp16")]; bool matmul_375_transpose_x_0 = const()[name = string("matmul_375_transpose_x_0"), val = bool(false)]; bool matmul_375_transpose_y_0 = const()[name = string("matmul_375_transpose_y_0"), val = bool(false)]; tensor matmul_375_cast_fp16 = matmul(transpose_x = matmul_375_transpose_x_0, transpose_y = matmul_375_transpose_y_0, x = reshape_1125_cast_fp16, y = reshape_1126_cast_fp16)[name = string("matmul_375_cast_fp16")]; tensor concat_3759 = const()[name = string("concat_3759"), val = tensor([1, 1, 104, 104])]; tensor reshape_1127_cast_fp16 = reshape(shape = concat_3759, x = matmul_375_cast_fp16)[name = string("reshape_1127_cast_fp16")]; tensor transpose_3063_perm_0 = const()[name = string("transpose_3063_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3063 = transpose(perm = transpose_3063_perm_0, x = reshape_1127_cast_fp16)[name = string("transpose_3096")]; tensor w_1503_cast_fp16 = add(x = transpose_3063, y = transpose_2305)[name = string("w_1503_cast_fp16")]; tensor var_7217_cast_fp16 = softmax(axis = var_7049, x = w_1503_cast_fp16)[name = string("op_7217_cast_fp16")]; string var_7219_equation_0 = const()[name = string("op_7219_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7219_cast_fp16 = einsum(equation = var_7219_equation_0, values = (var_7139_cast_fp16_7, var_7217_cast_fp16))[name = string("op_7219_cast_fp16")]; tensor transpose_752_perm_0 = const()[name = string("transpose_752_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3764 = const()[name = string("concat_3764"), val = tensor([1, 104, 64])]; tensor transpose_752_cast_fp16 = transpose(perm = transpose_752_perm_0, x = var_7105_cast_fp16_8)[name = string("transpose_3095")]; tensor reshape_1128_cast_fp16 = reshape(shape = concat_3764, x = transpose_752_cast_fp16)[name = string("reshape_1128_cast_fp16")]; tensor transpose_753_perm_0 = const()[name = string("transpose_753_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3765 = const()[name = string("concat_3765"), val = tensor([1, 64, 104])]; tensor transpose_753_cast_fp16 = transpose(perm = transpose_753_perm_0, x = var_7122_cast_fp16_8)[name = string("transpose_3094")]; tensor reshape_1129_cast_fp16 = reshape(shape = concat_3765, x = transpose_753_cast_fp16)[name = string("reshape_1129_cast_fp16")]; bool matmul_376_transpose_x_0 = const()[name = string("matmul_376_transpose_x_0"), val = bool(false)]; bool matmul_376_transpose_y_0 = const()[name = string("matmul_376_transpose_y_0"), val = bool(false)]; tensor matmul_376_cast_fp16 = matmul(transpose_x = matmul_376_transpose_x_0, transpose_y = matmul_376_transpose_y_0, x = reshape_1128_cast_fp16, y = reshape_1129_cast_fp16)[name = string("matmul_376_cast_fp16")]; tensor concat_3769 = const()[name = string("concat_3769"), val = tensor([1, 1, 104, 104])]; tensor reshape_1130_cast_fp16 = reshape(shape = concat_3769, x = matmul_376_cast_fp16)[name = string("reshape_1130_cast_fp16")]; tensor transpose_3064_perm_0 = const()[name = string("transpose_3064_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3064 = transpose(perm = transpose_3064_perm_0, x = reshape_1130_cast_fp16)[name = string("transpose_3093")]; tensor w_1507_cast_fp16 = add(x = transpose_3064, y = transpose_2305)[name = string("w_1507_cast_fp16")]; tensor var_7225_cast_fp16 = softmax(axis = var_7049, x = w_1507_cast_fp16)[name = string("op_7225_cast_fp16")]; string var_7227_equation_0 = const()[name = string("op_7227_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7227_cast_fp16 = einsum(equation = var_7227_equation_0, values = (var_7139_cast_fp16_8, var_7225_cast_fp16))[name = string("op_7227_cast_fp16")]; tensor transpose_754_perm_0 = const()[name = string("transpose_754_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3774 = const()[name = string("concat_3774"), val = tensor([1, 104, 64])]; tensor transpose_754_cast_fp16 = transpose(perm = transpose_754_perm_0, x = var_7105_cast_fp16_9)[name = string("transpose_3092")]; tensor reshape_1131_cast_fp16 = reshape(shape = concat_3774, x = transpose_754_cast_fp16)[name = string("reshape_1131_cast_fp16")]; tensor transpose_755_perm_0 = const()[name = string("transpose_755_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3775 = const()[name = string("concat_3775"), val = tensor([1, 64, 104])]; tensor transpose_755_cast_fp16 = transpose(perm = transpose_755_perm_0, x = var_7122_cast_fp16_9)[name = string("transpose_3091")]; tensor reshape_1132_cast_fp16 = reshape(shape = concat_3775, x = transpose_755_cast_fp16)[name = string("reshape_1132_cast_fp16")]; bool matmul_377_transpose_x_0 = const()[name = string("matmul_377_transpose_x_0"), val = bool(false)]; bool matmul_377_transpose_y_0 = const()[name = string("matmul_377_transpose_y_0"), val = bool(false)]; tensor matmul_377_cast_fp16 = matmul(transpose_x = matmul_377_transpose_x_0, transpose_y = matmul_377_transpose_y_0, x = reshape_1131_cast_fp16, y = reshape_1132_cast_fp16)[name = string("matmul_377_cast_fp16")]; tensor concat_3779 = const()[name = string("concat_3779"), val = tensor([1, 1, 104, 104])]; tensor reshape_1133_cast_fp16 = reshape(shape = concat_3779, x = matmul_377_cast_fp16)[name = string("reshape_1133_cast_fp16")]; tensor transpose_3065_perm_0 = const()[name = string("transpose_3065_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3065 = transpose(perm = transpose_3065_perm_0, x = reshape_1133_cast_fp16)[name = string("transpose_3090")]; tensor w_1511_cast_fp16 = add(x = transpose_3065, y = transpose_2305)[name = string("w_1511_cast_fp16")]; tensor var_7233_cast_fp16 = softmax(axis = var_7049, x = w_1511_cast_fp16)[name = string("op_7233_cast_fp16")]; string var_7235_equation_0 = const()[name = string("op_7235_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7235_cast_fp16 = einsum(equation = var_7235_equation_0, values = (var_7139_cast_fp16_9, var_7233_cast_fp16))[name = string("op_7235_cast_fp16")]; tensor transpose_756_perm_0 = const()[name = string("transpose_756_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3784 = const()[name = string("concat_3784"), val = tensor([1, 104, 64])]; tensor transpose_756_cast_fp16 = transpose(perm = transpose_756_perm_0, x = var_7105_cast_fp16_10)[name = string("transpose_3089")]; tensor reshape_1134_cast_fp16 = reshape(shape = concat_3784, x = transpose_756_cast_fp16)[name = string("reshape_1134_cast_fp16")]; tensor transpose_757_perm_0 = const()[name = string("transpose_757_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3785 = const()[name = string("concat_3785"), val = tensor([1, 64, 104])]; tensor transpose_757_cast_fp16 = transpose(perm = transpose_757_perm_0, x = var_7122_cast_fp16_10)[name = string("transpose_3088")]; tensor reshape_1135_cast_fp16 = reshape(shape = concat_3785, x = transpose_757_cast_fp16)[name = string("reshape_1135_cast_fp16")]; bool matmul_378_transpose_x_0 = const()[name = string("matmul_378_transpose_x_0"), val = bool(false)]; bool matmul_378_transpose_y_0 = const()[name = string("matmul_378_transpose_y_0"), val = bool(false)]; tensor matmul_378_cast_fp16 = matmul(transpose_x = matmul_378_transpose_x_0, transpose_y = matmul_378_transpose_y_0, x = reshape_1134_cast_fp16, y = reshape_1135_cast_fp16)[name = string("matmul_378_cast_fp16")]; tensor concat_3789 = const()[name = string("concat_3789"), val = tensor([1, 1, 104, 104])]; tensor reshape_1136_cast_fp16 = reshape(shape = concat_3789, x = matmul_378_cast_fp16)[name = string("reshape_1136_cast_fp16")]; tensor transpose_3066_perm_0 = const()[name = string("transpose_3066_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3066 = transpose(perm = transpose_3066_perm_0, x = reshape_1136_cast_fp16)[name = string("transpose_3087")]; tensor w_1515_cast_fp16 = add(x = transpose_3066, y = transpose_2305)[name = string("w_1515_cast_fp16")]; tensor var_7241_cast_fp16 = softmax(axis = var_7049, x = w_1515_cast_fp16)[name = string("op_7241_cast_fp16")]; string var_7243_equation_0 = const()[name = string("op_7243_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7243_cast_fp16 = einsum(equation = var_7243_equation_0, values = (var_7139_cast_fp16_10, var_7241_cast_fp16))[name = string("op_7243_cast_fp16")]; tensor transpose_758_perm_0 = const()[name = string("transpose_758_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3794 = const()[name = string("concat_3794"), val = tensor([1, 104, 64])]; tensor transpose_758_cast_fp16 = transpose(perm = transpose_758_perm_0, x = var_7105_cast_fp16_11)[name = string("transpose_3086")]; tensor reshape_1137_cast_fp16 = reshape(shape = concat_3794, x = transpose_758_cast_fp16)[name = string("reshape_1137_cast_fp16")]; tensor transpose_759_perm_0 = const()[name = string("transpose_759_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3795 = const()[name = string("concat_3795"), val = tensor([1, 64, 104])]; tensor transpose_759_cast_fp16 = transpose(perm = transpose_759_perm_0, x = var_7122_cast_fp16_11)[name = string("transpose_3085")]; tensor reshape_1138_cast_fp16 = reshape(shape = concat_3795, x = transpose_759_cast_fp16)[name = string("reshape_1138_cast_fp16")]; bool matmul_379_transpose_x_0 = const()[name = string("matmul_379_transpose_x_0"), val = bool(false)]; bool matmul_379_transpose_y_0 = const()[name = string("matmul_379_transpose_y_0"), val = bool(false)]; tensor matmul_379_cast_fp16 = matmul(transpose_x = matmul_379_transpose_x_0, transpose_y = matmul_379_transpose_y_0, x = reshape_1137_cast_fp16, y = reshape_1138_cast_fp16)[name = string("matmul_379_cast_fp16")]; tensor concat_3799 = const()[name = string("concat_3799"), val = tensor([1, 1, 104, 104])]; tensor reshape_1139_cast_fp16 = reshape(shape = concat_3799, x = matmul_379_cast_fp16)[name = string("reshape_1139_cast_fp16")]; tensor transpose_3067_perm_0 = const()[name = string("transpose_3067_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3067 = transpose(perm = transpose_3067_perm_0, x = reshape_1139_cast_fp16)[name = string("transpose_3084")]; tensor w_1519_cast_fp16 = add(x = transpose_3067, y = transpose_2305)[name = string("w_1519_cast_fp16")]; tensor var_7249_cast_fp16 = softmax(axis = var_7049, x = w_1519_cast_fp16)[name = string("op_7249_cast_fp16")]; string var_7251_equation_0 = const()[name = string("op_7251_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7251_cast_fp16 = einsum(equation = var_7251_equation_0, values = (var_7139_cast_fp16_11, var_7249_cast_fp16))[name = string("op_7251_cast_fp16")]; tensor transpose_760_perm_0 = const()[name = string("transpose_760_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3804 = const()[name = string("concat_3804"), val = tensor([1, 104, 64])]; tensor transpose_760_cast_fp16 = transpose(perm = transpose_760_perm_0, x = var_7105_cast_fp16_12)[name = string("transpose_3083")]; tensor reshape_1140_cast_fp16 = reshape(shape = concat_3804, x = transpose_760_cast_fp16)[name = string("reshape_1140_cast_fp16")]; tensor transpose_761_perm_0 = const()[name = string("transpose_761_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3805 = const()[name = string("concat_3805"), val = tensor([1, 64, 104])]; tensor transpose_761_cast_fp16 = transpose(perm = transpose_761_perm_0, x = var_7122_cast_fp16_12)[name = string("transpose_3082")]; tensor reshape_1141_cast_fp16 = reshape(shape = concat_3805, x = transpose_761_cast_fp16)[name = string("reshape_1141_cast_fp16")]; bool matmul_380_transpose_x_0 = const()[name = string("matmul_380_transpose_x_0"), val = bool(false)]; bool matmul_380_transpose_y_0 = const()[name = string("matmul_380_transpose_y_0"), val = bool(false)]; tensor matmul_380_cast_fp16 = matmul(transpose_x = matmul_380_transpose_x_0, transpose_y = matmul_380_transpose_y_0, x = reshape_1140_cast_fp16, y = reshape_1141_cast_fp16)[name = string("matmul_380_cast_fp16")]; tensor concat_3809 = const()[name = string("concat_3809"), val = tensor([1, 1, 104, 104])]; tensor reshape_1142_cast_fp16 = reshape(shape = concat_3809, x = matmul_380_cast_fp16)[name = string("reshape_1142_cast_fp16")]; tensor transpose_3068_perm_0 = const()[name = string("transpose_3068_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3068 = transpose(perm = transpose_3068_perm_0, x = reshape_1142_cast_fp16)[name = string("transpose_3081")]; tensor w_1523_cast_fp16 = add(x = transpose_3068, y = transpose_2305)[name = string("w_1523_cast_fp16")]; tensor var_7257_cast_fp16 = softmax(axis = var_7049, x = w_1523_cast_fp16)[name = string("op_7257_cast_fp16")]; string var_7259_equation_0 = const()[name = string("op_7259_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7259_cast_fp16 = einsum(equation = var_7259_equation_0, values = (var_7139_cast_fp16_12, var_7257_cast_fp16))[name = string("op_7259_cast_fp16")]; tensor transpose_762_perm_0 = const()[name = string("transpose_762_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3814 = const()[name = string("concat_3814"), val = tensor([1, 104, 64])]; tensor transpose_762_cast_fp16 = transpose(perm = transpose_762_perm_0, x = var_7105_cast_fp16_13)[name = string("transpose_3080")]; tensor reshape_1143_cast_fp16 = reshape(shape = concat_3814, x = transpose_762_cast_fp16)[name = string("reshape_1143_cast_fp16")]; tensor transpose_763_perm_0 = const()[name = string("transpose_763_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3815 = const()[name = string("concat_3815"), val = tensor([1, 64, 104])]; tensor transpose_763_cast_fp16 = transpose(perm = transpose_763_perm_0, x = var_7122_cast_fp16_13)[name = string("transpose_3079")]; tensor reshape_1144_cast_fp16 = reshape(shape = concat_3815, x = transpose_763_cast_fp16)[name = string("reshape_1144_cast_fp16")]; bool matmul_381_transpose_x_0 = const()[name = string("matmul_381_transpose_x_0"), val = bool(false)]; bool matmul_381_transpose_y_0 = const()[name = string("matmul_381_transpose_y_0"), val = bool(false)]; tensor matmul_381_cast_fp16 = matmul(transpose_x = matmul_381_transpose_x_0, transpose_y = matmul_381_transpose_y_0, x = reshape_1143_cast_fp16, y = reshape_1144_cast_fp16)[name = string("matmul_381_cast_fp16")]; tensor concat_3819 = const()[name = string("concat_3819"), val = tensor([1, 1, 104, 104])]; tensor reshape_1145_cast_fp16 = reshape(shape = concat_3819, x = matmul_381_cast_fp16)[name = string("reshape_1145_cast_fp16")]; tensor transpose_3069_perm_0 = const()[name = string("transpose_3069_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3069 = transpose(perm = transpose_3069_perm_0, x = reshape_1145_cast_fp16)[name = string("transpose_3078")]; tensor w_1527_cast_fp16 = add(x = transpose_3069, y = transpose_2305)[name = string("w_1527_cast_fp16")]; tensor var_7265_cast_fp16 = softmax(axis = var_7049, x = w_1527_cast_fp16)[name = string("op_7265_cast_fp16")]; string var_7267_equation_0 = const()[name = string("op_7267_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7267_cast_fp16 = einsum(equation = var_7267_equation_0, values = (var_7139_cast_fp16_13, var_7265_cast_fp16))[name = string("op_7267_cast_fp16")]; tensor transpose_764_perm_0 = const()[name = string("transpose_764_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3824 = const()[name = string("concat_3824"), val = tensor([1, 104, 64])]; tensor transpose_764_cast_fp16 = transpose(perm = transpose_764_perm_0, x = var_7105_cast_fp16_14)[name = string("transpose_3077")]; tensor reshape_1146_cast_fp16 = reshape(shape = concat_3824, x = transpose_764_cast_fp16)[name = string("reshape_1146_cast_fp16")]; tensor transpose_765_perm_0 = const()[name = string("transpose_765_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3825 = const()[name = string("concat_3825"), val = tensor([1, 64, 104])]; tensor transpose_765_cast_fp16 = transpose(perm = transpose_765_perm_0, x = var_7122_cast_fp16_14)[name = string("transpose_3076")]; tensor reshape_1147_cast_fp16 = reshape(shape = concat_3825, x = transpose_765_cast_fp16)[name = string("reshape_1147_cast_fp16")]; bool matmul_382_transpose_x_0 = const()[name = string("matmul_382_transpose_x_0"), val = bool(false)]; bool matmul_382_transpose_y_0 = const()[name = string("matmul_382_transpose_y_0"), val = bool(false)]; tensor matmul_382_cast_fp16 = matmul(transpose_x = matmul_382_transpose_x_0, transpose_y = matmul_382_transpose_y_0, x = reshape_1146_cast_fp16, y = reshape_1147_cast_fp16)[name = string("matmul_382_cast_fp16")]; tensor concat_3829 = const()[name = string("concat_3829"), val = tensor([1, 1, 104, 104])]; tensor reshape_1148_cast_fp16 = reshape(shape = concat_3829, x = matmul_382_cast_fp16)[name = string("reshape_1148_cast_fp16")]; tensor transpose_3070_perm_0 = const()[name = string("transpose_3070_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3070 = transpose(perm = transpose_3070_perm_0, x = reshape_1148_cast_fp16)[name = string("transpose_3075")]; tensor w_1531_cast_fp16 = add(x = transpose_3070, y = transpose_2305)[name = string("w_1531_cast_fp16")]; tensor var_7273_cast_fp16 = softmax(axis = var_7049, x = w_1531_cast_fp16)[name = string("op_7273_cast_fp16")]; string var_7275_equation_0 = const()[name = string("op_7275_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7275_cast_fp16 = einsum(equation = var_7275_equation_0, values = (var_7139_cast_fp16_14, var_7273_cast_fp16))[name = string("op_7275_cast_fp16")]; tensor transpose_766_perm_0 = const()[name = string("transpose_766_perm_0"), val = tensor([0, 2, 3, 1])]; tensor concat_3834 = const()[name = string("concat_3834"), val = tensor([1, 104, 64])]; tensor transpose_766_cast_fp16 = transpose(perm = transpose_766_perm_0, x = var_7105_cast_fp16_15)[name = string("transpose_3074")]; tensor reshape_1149_cast_fp16 = reshape(shape = concat_3834, x = transpose_766_cast_fp16)[name = string("reshape_1149_cast_fp16")]; tensor transpose_767_perm_0 = const()[name = string("transpose_767_perm_0"), val = tensor([0, 2, 1, 3])]; tensor concat_3835 = const()[name = string("concat_3835"), val = tensor([1, 64, 104])]; tensor transpose_767_cast_fp16 = transpose(perm = transpose_767_perm_0, x = var_7122_cast_fp16_15)[name = string("transpose_3073")]; tensor reshape_1150_cast_fp16 = reshape(shape = concat_3835, x = transpose_767_cast_fp16)[name = string("reshape_1150_cast_fp16")]; bool matmul_383_transpose_x_0 = const()[name = string("matmul_383_transpose_x_0"), val = bool(false)]; bool matmul_383_transpose_y_0 = const()[name = string("matmul_383_transpose_y_0"), val = bool(false)]; tensor matmul_383_cast_fp16 = matmul(transpose_x = matmul_383_transpose_x_0, transpose_y = matmul_383_transpose_y_0, x = reshape_1149_cast_fp16, y = reshape_1150_cast_fp16)[name = string("matmul_383_cast_fp16")]; tensor concat_3839 = const()[name = string("concat_3839"), val = tensor([1, 1, 104, 104])]; tensor reshape_1151_cast_fp16 = reshape(shape = concat_3839, x = matmul_383_cast_fp16)[name = string("reshape_1151_cast_fp16")]; tensor transpose_3071_perm_0 = const()[name = string("transpose_3071_perm_0"), val = tensor([0, 3, 1, 2])]; tensor transpose_3071 = transpose(perm = transpose_3071_perm_0, x = reshape_1151_cast_fp16)[name = string("transpose_3072")]; tensor w_cast_fp16 = add(x = transpose_3071, y = transpose_2305)[name = string("w_cast_fp16")]; tensor var_7281_cast_fp16 = softmax(axis = var_7049, x = w_cast_fp16)[name = string("op_7281_cast_fp16")]; string var_7283_equation_0 = const()[name = string("op_7283_equation_0"), val = string("bchk,bkhq->bchq")]; tensor var_7283_cast_fp16 = einsum(equation = var_7283_equation_0, values = (var_7139_cast_fp16_15, var_7281_cast_fp16))[name = string("op_7283_cast_fp16")]; bool input_195_interleave_0 = const()[name = string("input_195_interleave_0"), val = bool(false)]; tensor input_195_cast_fp16 = concat(axis = var_7049, interleave = input_195_interleave_0, values = (var_7163_cast_fp16, var_7171_cast_fp16, var_7179_cast_fp16, var_7187_cast_fp16, var_7195_cast_fp16, var_7203_cast_fp16, var_7211_cast_fp16, var_7219_cast_fp16, var_7227_cast_fp16, var_7235_cast_fp16, var_7243_cast_fp16, var_7251_cast_fp16, var_7259_cast_fp16, var_7267_cast_fp16, var_7275_cast_fp16, var_7283_cast_fp16))[name = string("input_195_cast_fp16")]; string var_7292_pad_type_0 = const()[name = string("op_7292_pad_type_0"), val = string("valid")]; tensor var_7292_strides_0 = const()[name = string("op_7292_strides_0"), val = tensor([1, 1])]; tensor var_7292_pad_0 = const()[name = string("op_7292_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_7292_dilations_0 = const()[name = string("op_7292_dilations_0"), val = tensor([1, 1])]; int32 var_7292_groups_0 = const()[name = string("op_7292_groups_0"), val = int32(1)]; tensor layers_23_self_attn_out_proj_weight_to_fp16 = const()[name = string("layers_23_self_attn_out_proj_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(609827264)))]; tensor layers_23_self_attn_out_proj_bias_to_fp16 = const()[name = string("layers_23_self_attn_out_proj_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(611924480)))]; tensor var_7292_cast_fp16 = conv(bias = layers_23_self_attn_out_proj_bias_to_fp16, dilations = var_7292_dilations_0, groups = var_7292_groups_0, pad = var_7292_pad_0, pad_type = var_7292_pad_type_0, strides = var_7292_strides_0, weight = layers_23_self_attn_out_proj_weight_to_fp16, x = input_195_cast_fp16)[name = string("op_7292_cast_fp16")]; tensor x_247_cast_fp16 = add(x = x_243_cast_fp16, y = var_7292_cast_fp16)[name = string("x_247_cast_fp16")]; tensor mu_95_axes_0 = const()[name = string("mu_95_axes_0"), val = tensor([1])]; bool mu_95_keep_dims_0 = const()[name = string("mu_95_keep_dims_0"), val = bool(true)]; tensor mu_95_cast_fp16 = reduce_mean(axes = mu_95_axes_0, keep_dims = mu_95_keep_dims_0, x = x_247_cast_fp16)[name = string("mu_95_cast_fp16")]; tensor var_7298_cast_fp16 = sub(x = x_247_cast_fp16, y = mu_95_cast_fp16)[name = string("op_7298_cast_fp16")]; fp16 var_7052_promoted_1_to_fp16 = const()[name = string("op_7052_promoted_1_to_fp16"), val = fp16(0x1p+1)]; tensor var_7299_cast_fp16 = pow(x = var_7298_cast_fp16, y = var_7052_promoted_1_to_fp16)[name = string("op_7299_cast_fp16")]; tensor var_95_axes_0 = const()[name = string("var_95_axes_0"), val = tensor([1])]; bool var_95_keep_dims_0 = const()[name = string("var_95_keep_dims_0"), val = bool(true)]; tensor var_95_cast_fp16 = reduce_mean(axes = var_95_axes_0, keep_dims = var_95_keep_dims_0, x = var_7299_cast_fp16)[name = string("var_95_cast_fp16")]; fp16 var_7303_to_fp16 = const()[name = string("op_7303_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_7304_cast_fp16 = add(x = var_95_cast_fp16, y = var_7303_to_fp16)[name = string("op_7304_cast_fp16")]; fp32 var_7305_epsilon_0 = const()[name = string("op_7305_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_7305_cast_fp16 = rsqrt(epsilon = var_7305_epsilon_0, x = var_7304_cast_fp16)[name = string("op_7305_cast_fp16")]; tensor x_249_cast_fp16 = mul(x = var_7298_cast_fp16, y = var_7305_cast_fp16)[name = string("x_249_cast_fp16")]; tensor input_197_gamma_0_to_fp16 = const()[name = string("input_197_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(611926592)))]; tensor input_197_beta_0_to_fp16 = const()[name = string("input_197_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(611928704)))]; fp16 input_197_epsilon_0_to_fp16 = const()[name = string("input_197_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_197_cast_fp16 = batch_norm(beta = input_197_beta_0_to_fp16, epsilon = input_197_epsilon_0_to_fp16, gamma = input_197_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_249_cast_fp16)[name = string("input_197_cast_fp16")]; string x_251_pad_type_0 = const()[name = string("x_251_pad_type_0"), val = string("valid")]; tensor x_251_strides_0 = const()[name = string("x_251_strides_0"), val = tensor([1, 1])]; tensor x_251_pad_0 = const()[name = string("x_251_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_251_dilations_0 = const()[name = string("x_251_dilations_0"), val = tensor([1, 1])]; int32 x_251_groups_0 = const()[name = string("x_251_groups_0"), val = int32(1)]; tensor layers_23_fc1_weight_to_fp16 = const()[name = string("layers_23_fc1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(611930816)))]; tensor layers_23_fc1_bias_to_fp16 = const()[name = string("layers_23_fc1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(620319488)))]; tensor x_251_cast_fp16 = conv(bias = layers_23_fc1_bias_to_fp16, dilations = x_251_dilations_0, groups = x_251_groups_0, pad = x_251_pad_0, pad_type = x_251_pad_type_0, strides = x_251_strides_0, weight = layers_23_fc1_weight_to_fp16, x = input_197_cast_fp16)[name = string("x_251_cast_fp16")]; fp16 var_7320_to_fp16 = const()[name = string("op_7320_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_7321_cast_fp16 = mul(x = x_251_cast_fp16, y = var_7320_to_fp16)[name = string("op_7321_cast_fp16")]; tensor var_7322_cast_fp16 = mul(x = var_7321_cast_fp16, y = x_251_cast_fp16)[name = string("op_7322_cast_fp16")]; tensor var_7323_cast_fp16 = mul(x = var_7322_cast_fp16, y = x_251_cast_fp16)[name = string("op_7323_cast_fp16")]; tensor var_7324_cast_fp16 = add(x = x_251_cast_fp16, y = var_7323_cast_fp16)[name = string("op_7324_cast_fp16")]; fp16 var_7325_to_fp16 = const()[name = string("op_7325_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_53_cast_fp16 = mul(x = var_7324_cast_fp16, y = var_7325_to_fp16)[name = string("u_53_cast_fp16")]; fp16 var_7327_to_fp16 = const()[name = string("op_7327_to_fp16"), val = fp16(0x1p-1)]; tensor var_7328_cast_fp16 = mul(x = x_251_cast_fp16, y = var_7327_to_fp16)[name = string("op_7328_cast_fp16")]; tensor var_7329_cast_fp16 = tanh(x = u_53_cast_fp16)[name = string("op_7329_cast_fp16")]; fp16 var_7330_to_fp16 = const()[name = string("op_7330_to_fp16"), val = fp16(0x1p+0)]; tensor var_7331_cast_fp16 = add(x = var_7329_cast_fp16, y = var_7330_to_fp16)[name = string("op_7331_cast_fp16")]; tensor input_199_cast_fp16 = mul(x = var_7328_cast_fp16, y = var_7331_cast_fp16)[name = string("input_199_cast_fp16")]; string h_pad_type_0 = const()[name = string("h_pad_type_0"), val = string("valid")]; tensor h_strides_0 = const()[name = string("h_strides_0"), val = tensor([1, 1])]; tensor h_pad_0 = const()[name = string("h_pad_0"), val = tensor([0, 0, 0, 0])]; tensor h_dilations_0 = const()[name = string("h_dilations_0"), val = tensor([1, 1])]; int32 h_groups_0 = const()[name = string("h_groups_0"), val = int32(1)]; tensor layers_23_fc2_weight_to_fp16 = const()[name = string("layers_23_fc2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(620327744)))]; tensor layers_23_fc2_bias_to_fp16 = const()[name = string("layers_23_fc2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(628716416)))]; tensor h_cast_fp16 = conv(bias = layers_23_fc2_bias_to_fp16, dilations = h_dilations_0, groups = h_groups_0, pad = h_pad_0, pad_type = h_pad_type_0, strides = h_strides_0, weight = layers_23_fc2_weight_to_fp16, x = input_199_cast_fp16)[name = string("h_cast_fp16")]; tensor x_253_cast_fp16 = add(x = x_247_cast_fp16, y = h_cast_fp16)[name = string("x_253_cast_fp16")]; tensor mu_axes_0 = const()[name = string("mu_axes_0"), val = tensor([1])]; bool mu_keep_dims_0 = const()[name = string("mu_keep_dims_0"), val = bool(true)]; tensor mu_cast_fp16 = reduce_mean(axes = mu_axes_0, keep_dims = mu_keep_dims_0, x = x_253_cast_fp16)[name = string("mu_cast_fp16")]; tensor var_7350_cast_fp16 = sub(x = x_253_cast_fp16, y = mu_cast_fp16)[name = string("op_7350_cast_fp16")]; fp16 var_7342_promoted_to_fp16 = const()[name = string("op_7342_promoted_to_fp16"), val = fp16(0x1p+1)]; tensor var_7351_cast_fp16 = pow(x = var_7350_cast_fp16, y = var_7342_promoted_to_fp16)[name = string("op_7351_cast_fp16")]; tensor var_axes_0 = const()[name = string("var_axes_0"), val = tensor([1])]; bool var_keep_dims_0 = const()[name = string("var_keep_dims_0"), val = bool(true)]; tensor var_cast_fp16 = reduce_mean(axes = var_axes_0, keep_dims = var_keep_dims_0, x = var_7351_cast_fp16)[name = string("var_cast_fp16")]; fp16 var_7355_to_fp16 = const()[name = string("op_7355_to_fp16"), val = fp16(0x1.5p-17)]; tensor var_7356_cast_fp16 = add(x = var_cast_fp16, y = var_7355_to_fp16)[name = string("op_7356_cast_fp16")]; fp32 var_7357_epsilon_0 = const()[name = string("op_7357_epsilon_0"), val = fp32(0x1.197998p-40)]; tensor var_7357_cast_fp16 = rsqrt(epsilon = var_7357_epsilon_0, x = var_7356_cast_fp16)[name = string("op_7357_cast_fp16")]; tensor x_255_cast_fp16 = mul(x = var_7350_cast_fp16, y = var_7357_cast_fp16)[name = string("x_255_cast_fp16")]; tensor input_201_gamma_0_to_fp16 = const()[name = string("input_201_gamma_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(628718528)))]; tensor input_201_beta_0_to_fp16 = const()[name = string("input_201_beta_0_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(628720640)))]; fp16 input_201_epsilon_0_to_fp16 = const()[name = string("input_201_epsilon_0_to_fp16"), val = fp16(0x1.5p-17)]; tensor input_201_cast_fp16 = batch_norm(beta = input_201_beta_0_to_fp16, epsilon = input_201_epsilon_0_to_fp16, gamma = input_201_gamma_0_to_fp16, mean = input_9_mean_0_to_fp16, variance = input_9_variance_0_to_fp16, x = x_255_cast_fp16)[name = string("input_201_cast_fp16")]; string x_pad_type_0 = const()[name = string("x_pad_type_0"), val = string("valid")]; tensor x_strides_0 = const()[name = string("x_strides_0"), val = tensor([1, 1])]; tensor x_pad_0 = const()[name = string("x_pad_0"), val = tensor([0, 0, 0, 0])]; tensor x_dilations_0 = const()[name = string("x_dilations_0"), val = tensor([1, 1])]; int32 x_groups_0 = const()[name = string("x_groups_0"), val = int32(1)]; tensor proj1_weight_to_fp16 = const()[name = string("proj1_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(628722752)))]; tensor proj1_bias_to_fp16 = const()[name = string("proj1_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(630819968)))]; tensor x_cast_fp16 = conv(bias = proj1_bias_to_fp16, dilations = x_dilations_0, groups = x_groups_0, pad = x_pad_0, pad_type = x_pad_type_0, strides = x_strides_0, weight = proj1_weight_to_fp16, x = input_201_cast_fp16)[name = string("x_cast_fp16")]; fp16 var_7376_to_fp16 = const()[name = string("op_7376_to_fp16"), val = fp16(0x1.6e4p-5)]; tensor var_7377_cast_fp16 = mul(x = x_cast_fp16, y = var_7376_to_fp16)[name = string("op_7377_cast_fp16")]; tensor var_7378_cast_fp16 = mul(x = var_7377_cast_fp16, y = x_cast_fp16)[name = string("op_7378_cast_fp16")]; tensor var_7379_cast_fp16 = mul(x = var_7378_cast_fp16, y = x_cast_fp16)[name = string("op_7379_cast_fp16")]; tensor var_7381_cast_fp16 = add(x = x_cast_fp16, y = var_7379_cast_fp16)[name = string("op_7381_cast_fp16")]; fp16 var_7382_to_fp16 = const()[name = string("op_7382_to_fp16"), val = fp16(0x1.988p-1)]; tensor u_cast_fp16 = mul(x = var_7381_cast_fp16, y = var_7382_to_fp16)[name = string("u_cast_fp16")]; fp16 var_7384_to_fp16 = const()[name = string("op_7384_to_fp16"), val = fp16(0x1p-1)]; tensor var_7385_cast_fp16 = mul(x = x_cast_fp16, y = var_7384_to_fp16)[name = string("op_7385_cast_fp16")]; tensor var_7386_cast_fp16 = tanh(x = u_cast_fp16)[name = string("op_7386_cast_fp16")]; fp16 var_7388_to_fp16 = const()[name = string("op_7388_to_fp16"), val = fp16(0x1p+0)]; tensor var_7389_cast_fp16 = add(x = var_7386_cast_fp16, y = var_7388_to_fp16)[name = string("op_7389_cast_fp16")]; tensor input_cast_fp16 = mul(x = var_7385_cast_fp16, y = var_7389_cast_fp16)[name = string("input_cast_fp16")]; string var_7401_pad_type_0 = const()[name = string("op_7401_pad_type_0"), val = string("valid")]; tensor var_7401_strides_0 = const()[name = string("op_7401_strides_0"), val = tensor([1, 1])]; tensor var_7401_pad_0 = const()[name = string("op_7401_pad_0"), val = tensor([0, 0, 0, 0])]; tensor var_7401_dilations_0 = const()[name = string("op_7401_dilations_0"), val = tensor([1, 1])]; int32 var_7401_groups_0 = const()[name = string("op_7401_groups_0"), val = int32(1)]; tensor proj2_weight_to_fp16 = const()[name = string("proj2_weight_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(630822080)))]; tensor proj2_bias_to_fp16 = const()[name = string("proj2_bias_to_fp16"), val = tensor(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(635016448)))]; tensor audio_embeds = conv(bias = proj2_bias_to_fp16, dilations = var_7401_dilations_0, groups = var_7401_groups_0, pad = var_7401_pad_0, pad_type = var_7401_pad_type_0, strides = var_7401_strides_0, weight = proj2_weight_to_fp16, x = input_cast_fp16)[name = string("op_7401_cast_fp16")]; } -> (audio_embeds); }