Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,7 @@
+---
+language: en
+library_name: mlx
+pipeline_tag: text-generation
+tags:
+- mlx
+---

chat_template.jinja

@@ -0,0 +1,124 @@
+{%- if tools or xml_tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {%- if custom_instructions %}
+        {{- custom_instructions ~ '\n\n' }}
+    {%- endif -%}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in (xml_tools or tools) %}
+        {{- "\n" }}
+        {{- tool }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call>" }}
+    {%- if python_tools -%}
+        {{- python_tools -}}
+    {%- endif -%}
+    {{- "<|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content }}
+        {%- if custom_instructions %}
+            {{- custom_instructions -}}
+        {%- endif -%}
+        {{- '<|im_end|>\n' }}
+    {%- elif custom_instructions -%}
+        {{- '<|im_start|>system\n' ~ custom_instructions ~ '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {{- '<|im_start|>' + message.role + '\n' }}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if (not loop.last and reasoning_content) %}
+                {{- '/reasoning_' ~ (reasoning_lang | default('en', true)) ~ '\n' }}
+                {%- if enable_turbo %}
+                    {{- '/turbo\n' }}
+                {%- endif %}
+                {% generation %}
+                {{- '<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') -}}
+                {% endgeneration %}
+            {%- elif loop.last %}
+                {%- if reasoning_content %}
+                    {{- '/reasoning_' ~ (reasoning_lang | default('en', true)) ~ '\n' }}
+                    {%- if enable_turbo %}
+                        {{- '/turbo\n' }}
+                    {%- endif %}
+                    {% generation %}
+                    {{- '<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') -}}
+                    {% endgeneration %}
+                {%- else %}
+                    {{- '<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' }}
+                    {% generation %}
+                    {{- content.lstrip('\n') -}}
+                    {% endgeneration %}
+                {%- endif %}
+            {%- else %}
+                {% generation %}
+                {{- content -}}
+                {% endgeneration %}
+            {%- endif %}
+        {%- else %}
+            {% generation %}
+            {{- content -}}
+            {% endgeneration %}
+        {%- endif %}
+        {% generation %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                 {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {{- '<tool_call>\n' ~ tool_call ~ '\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>' }}
+        {% endgeneration %}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '/reasoning_' ~ (reasoning_lang | default('en', true)) ~ '\n' }}
+        {%- if enable_turbo %}
+            {{- '/turbo\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,56 @@
+{
+    "architectures": [
+        "EngGPTMoeForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "auto_map": {
+        "AutoConfig": "configuration_enggpt_moe.EngGPTMoeConfig",
+        "AutoModelForCausalLM": "modeling_enggpt_moe.EngGPTMoeForCausalLM"
+    },
+    "bos_token_id": 1,
+    "decoder_sparse_step": 1,
+    "dtype": "bfloat16",
+    "eos_token_id": 131073,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 2880,
+    "initializer_range": 0.02,
+    "intermediate_size": 8640,
+    "max_position_embeddings": 40960,
+    "max_window_layers": 24,
+    "mlp_only_layers": [],
+    "model_type": "enggpt_moe",
+    "moe_intermediate_size": 1080,
+    "norm_topk_prob": true,
+    "num_attention_heads": 32,
+    "num_experts": 64,
+    "num_experts_per_tok": 8,
+    "num_hidden_layers": 24,
+    "num_key_value_heads": 4,
+    "out_proj_bias": null,
+    "output_router_logits": false,
+    "pad_token_id": 131083,
+    "qkv_proj_bias": null,
+    "quantization": {
+        "group_size": 64,
+        "bits": 4,
+        "mode": "affine"
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 4,
+        "mode": "affine"
+    },
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 1000000.0,
+    "router_aux_loss_coef": 0.001,
+    "sliding_window": null,
+    "tie_word_embeddings": false,
+    "transformers_version": "4.57.0",
+    "use_bias": false,
+    "use_cache": false,
+    "use_sliding_window": false,
+    "vocab_size": 131084
+}

configuration_enggpt_moe.py

@@ -0,0 +1,98 @@
+"""EngGPT model configuration"""
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.modeling_rope_utils import rope_config_validation
+
+class EngGPTMoeConfig(PretrainedConfig):
+   
+    model_type = "enggpt_moe"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    base_model_tp_plan = {
+        "layers.*.self_attn.q_proj": "colwise",
+        "layers.*.self_attn.k_proj": "colwise",
+        "layers.*.self_attn.v_proj": "colwise",
+        "layers.*.self_attn.o_proj": "rowwise",
+        "layers.*.mlp.experts.*.gate_proj": "colwise",
+        "layers.*.mlp.experts.*.up_proj": "colwise",
+        "layers.*.mlp.experts.*.down_proj": "rowwise",
+        "layers.*.mlp.gate_proj": "colwise",
+        "layers.*.mlp.up_proj": "colwise",
+        "layers.*.mlp.down_proj": "rowwise",
+    }
+    base_model_pp_plan = {
+        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
+        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
+        "norm": (["hidden_states"], ["hidden_states"]),
+    }
+
+    def __init__(
+        self,
+        vocab_size=131084,
+        hidden_size=2880,
+        intermediate_size=8640,
+        num_hidden_layers=24,
+        num_attention_heads=32,
+        num_key_value_heads=4,
+        hidden_act="silu",
+        max_position_embeddings=32768,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=False,
+        tie_word_embeddings=False,
+        rope_theta=1000000.0,
+        rope_scaling=None,
+        attention_bias=False,
+        use_sliding_window=False,
+        sliding_window=None,
+        attention_dropout=0.0,
+        decoder_sparse_step=1,
+        moe_intermediate_size=1080,
+        num_experts_per_tok=8,
+        num_experts=64,
+        norm_topk_prob=True,
+        output_router_logits=False,
+        router_aux_loss_coef=0.001,
+        mlp_only_layers=None,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.use_sliding_window = use_sliding_window
+        self.sliding_window = sliding_window if use_sliding_window else None
+
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+        
+        if self.rope_scaling is not None and "type" in self.rope_scaling:
+            self.rope_scaling["rope_type"] = self.rope_scaling["type"]
+        rope_config_validation(self)
+
+        # MoE arguments
+        self.decoder_sparse_step = decoder_sparse_step
+        self.moe_intermediate_size = moe_intermediate_size
+        self.num_experts_per_tok = num_experts_per_tok
+        self.num_experts = num_experts
+        self.norm_topk_prob = norm_topk_prob
+        self.output_router_logits = output_router_logits
+        self.router_aux_loss_coef = router_aux_loss_coef
+        self.mlp_only_layers = [] if mlp_only_layers is None else mlp_only_layers
+
+        super().__init__(
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+
+
+__all__ = ["EngGPTMoeConfig"]

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 131073,
+  "pad_token_id": 131083,
+  "transformers_version": "4.57.6"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db82e121b88af8ef962d46c06b61d01724fe4a8e0d315219a8060baa036893a1
+size 5324421483

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b46506c4c36197ce0af1525a6bfae063ba67da9c2754b767a73083fd95a7047
+size 5320978891

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3eddd192aa937088a68d93378009be5b93a9a0c3a6803407bc454238282c5ea
+size 5070405996

model.safetensors.index.json

@@ -0,0 +1,639 @@
+{
+    "metadata": {
+        "total_size": 15715733088,
+        "total_parameters": 15729347904
+    },
+    "weight_map": {
+        "lm_head.biases": "model-00003-of-00003.safetensors",
+        "lm_head.scales": "model-00003-of-00003.safetensors",
+        "lm_head.weight": "model-00003-of-00003.safetensors",
+        "model.embed_tokens.biases": "model-00001-of-00003.safetensors",
+        "model.embed_tokens.scales": "model-00001-of-00003.safetensors",
+        "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.16.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.16.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.17.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.17.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.20.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.gate.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.gate.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.gate.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.gate_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.gate_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.up_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.up_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.mlp.switch_mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.k_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.o_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.q_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.v_proj.biases": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
+        "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+        "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.gate_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.gate_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.up_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.up_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.mlp.switch_mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.gate.biases": "model-00001-of-00003.safetensors",
+        "model.layers.8.mlp.gate.scales": "model-00001-of-00003.safetensors",
+        "model.layers.8.mlp.gate.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.8.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.8.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.k_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.o_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.q_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.v_proj.biases": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
+        "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+        "model.layers.9.input_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.gate.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.gate.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.gate.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.gate_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.gate_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.up_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.up_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.mlp.switch_mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.k_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.o_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.q_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.v_proj.biases": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
+        "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+        "model.norm.weight": "model-00003-of-00003.safetensors"
+    }
+}

modeling_enggpt_moe.py

@@ -0,0 +1,688 @@
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+#           This file was automatically generated from src/transformers/models/enggpt_moe/modular_enggpt_moe.py.
+#               Do NOT edit this file manually as any edits will be overwritten by the generation of
+#             the file from the modular. If any change should be done, please apply the change to the
+#                          modular_enggpt_moe.py file directly. One of our CI enforces this.
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+
+from typing import Callable, Optional, Union
+
+import torch
+import torch.nn.functional as F
+from torch import nn
+
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache
+from transformers.generation import GenerationMixin
+from transformers.integrations import use_kernel_forward_from_hub
+from transformers.masking_utils import create_causal_mask, create_sliding_window_causal_mask
+from transformers.modeling_flash_attention_utils import FlashAttentionKwargs
+from transformers.modeling_layers import (
+    GenericForQuestionAnswering,
+    GenericForSequenceClassification,
+    GenericForTokenClassification,
+    GradientCheckpointingLayer,
+)
+from transformers.modeling_outputs import MoeCausalLMOutputWithPast, MoeModelOutputWithPast
+from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
+from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
+from transformers.processing_utils import Unpack
+from transformers.utils import TransformersKwargs, auto_docstring, can_return_tuple
+from transformers.utils.deprecation import deprecate_kwarg
+from transformers.utils.generic import OutputRecorder, check_model_inputs
+from .configuration_enggpt_moe import EngGPTMoeConfig
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+
+def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1):
+    """Applies Rotary Position Embedding to the query and key tensors.
+
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+        position_ids (`torch.Tensor`, *optional*):
+            Deprecated and unused.
+        unsqueeze_dim (`int`, *optional*, defaults to 1):
+            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
+            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
+            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
+            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
+            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
+            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
+    Returns:
+        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    cos = cos.unsqueeze(unsqueeze_dim)
+    sin = sin.unsqueeze(unsqueeze_dim)
+    q_embed = (q * cos) + (rotate_half(q) * sin)
+    k_embed = (k * cos) + (rotate_half(k) * sin)
+    return q_embed, k_embed
+
+
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+def eager_attention_forward(
+    module: nn.Module,
+    query: torch.Tensor,
+    key: torch.Tensor,
+    value: torch.Tensor,
+    attention_mask: Optional[torch.Tensor],
+    scaling: float,
+    dropout: float = 0.0,
+    **kwargs: Unpack[TransformersKwargs],
+):
+    key_states = repeat_kv(key, module.num_key_value_groups)
+    value_states = repeat_kv(value, module.num_key_value_groups)
+
+    attn_weights = torch.matmul(query, key_states.transpose(2, 3)) * scaling
+    if attention_mask is not None:
+        causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]
+        attn_weights = attn_weights + causal_mask
+
+    attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query.dtype)
+    attn_weights = nn.functional.dropout(attn_weights, p=dropout, training=module.training)
+    attn_output = torch.matmul(attn_weights, value_states)
+    attn_output = attn_output.transpose(1, 2).contiguous()
+
+    return attn_output, attn_weights
+
+
+class EngGPTMoeAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config: EngGPTMoeConfig, layer_idx: int):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        self.head_dim = getattr(config, "head_dim", config.hidden_size // config.num_attention_heads)
+        self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads
+        self.scaling = self.head_dim**-0.5
+        self.attention_dropout = config.attention_dropout
+        self.is_causal = True
+
+        self.q_proj = nn.Linear(
+            config.hidden_size, config.num_attention_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.k_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.v_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.o_proj = nn.Linear(
+            config.num_attention_heads * self.head_dim, config.hidden_size, bias=config.attention_bias
+        )
+        self.q_norm = EngGPTMoeRMSNorm(self.head_dim, eps=config.rms_norm_eps)  # unlike olmo, only on the head dim!
+        self.k_norm = EngGPTMoeRMSNorm(
+            self.head_dim, eps=config.rms_norm_eps
+        )  # thus post q_norm does not need reshape
+        self.sliding_window = getattr(config, "sliding_window", None)
+
+    @deprecate_kwarg("past_key_value", new_name="past_key_values", version="4.58")
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        position_embeddings: tuple[torch.Tensor, torch.Tensor],
+        attention_mask: Optional[torch.Tensor],
+        past_key_values: Optional[Cache] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> tuple[torch.Tensor, Optional[torch.Tensor]]:
+        input_shape = hidden_states.shape[:-1]
+        hidden_shape = (*input_shape, -1, self.head_dim)
+
+        query_states = self.q_norm(self.q_proj(hidden_states).view(hidden_shape)).transpose(1, 2)
+        key_states = self.k_norm(self.k_proj(hidden_states).view(hidden_shape)).transpose(1, 2)
+        value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+
+        cos, sin = position_embeddings
+        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
+
+        if past_key_values is not None:
+            # sin and cos are specific to RoPE models; cache_position needed for the static cache
+            cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
+            key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        attention_interface: Callable = eager_attention_forward
+        if self.config._attn_implementation != "eager":
+            attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation]
+
+        attn_output, attn_weights = attention_interface(
+            self,
+            query_states,
+            key_states,
+            value_states,
+            attention_mask,
+            dropout=0.0 if not self.training else self.attention_dropout,
+            scaling=self.scaling,
+            sliding_window=self.sliding_window,  # diff with Llama
+            **kwargs,
+        )
+
+        attn_output = attn_output.reshape(*input_shape, -1).contiguous()
+        attn_output = self.o_proj(attn_output)
+        return attn_output, attn_weights
+
+
+class EngGPTMoeMLP(nn.Module):
+    def __init__(self, config, intermediate_size=None):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = intermediate_size if intermediate_size is not None else config.intermediate_size
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.act_fn = ACT2FN[config.hidden_act]
+
+    def forward(self, x):
+        down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+        return down_proj
+
+
+class EngGPTMoeSparseMoeBlock(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.num_experts = config.num_experts
+        self.top_k = config.num_experts_per_tok
+        self.norm_topk_prob = config.norm_topk_prob
+
+        # gating
+        self.gate = nn.Linear(config.hidden_size, config.num_experts, bias=False)
+        self.experts = nn.ModuleList(
+            [EngGPTMoeMLP(config, intermediate_size=config.moe_intermediate_size) for _ in range(self.num_experts)]
+        )
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        batch_size, sequence_length, hidden_dim = hidden_states.shape
+        hidden_states = hidden_states.view(-1, hidden_dim)
+
+        router_logits = self.gate(hidden_states)
+
+        routing_weights = F.softmax(router_logits, dim=1, dtype=torch.float)
+        routing_weights, selected_experts = torch.topk(routing_weights, self.top_k, dim=-1)
+        if self.norm_topk_prob:
+            routing_weights /= routing_weights.sum(dim=-1, keepdim=True)
+
+        routing_weights = routing_weights.to(hidden_states.dtype)
+
+        final_hidden_states = torch.zeros(
+            (batch_size * sequence_length, hidden_dim), dtype=hidden_states.dtype, device=hidden_states.device
+        )
+
+        expert_mask = torch.nn.functional.one_hot(selected_experts, num_classes=self.num_experts).permute(2, 1, 0)
+
+        expert_hit = torch.greater(expert_mask.sum(dim=(-1, -2)), 0).nonzero()
+        for expert_idx in expert_hit:
+            expert_layer = self.experts[expert_idx]
+            idx, top_x = torch.where(expert_mask[expert_idx].squeeze(0))
+
+            current_state = hidden_states[None, top_x].reshape(-1, hidden_dim)
+            current_hidden_states = expert_layer(current_state) * routing_weights[top_x, idx, None]
+
+            final_hidden_states.index_add_(0, top_x, current_hidden_states.to(hidden_states.dtype))
+        final_hidden_states = final_hidden_states.reshape(batch_size, sequence_length, hidden_dim)
+        return final_hidden_states, router_logits
+
+
+@use_kernel_forward_from_hub("RMSNorm")
+class EngGPTMoeRMSNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """
+        EngGPTMoeRMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, hidden_states):
+        input_dtype = hidden_states.dtype
+        hidden_states = hidden_states.to(torch.float32)
+        variance = hidden_states.pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
+        return self.weight * hidden_states.to(input_dtype)
+
+    def extra_repr(self):
+        return f"{tuple(self.weight.shape)}, eps={self.variance_epsilon}"
+
+
+class EngGPTMoeDecoderLayer(GradientCheckpointingLayer):
+    def __init__(self, config: EngGPTMoeConfig, layer_idx: int):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+
+        self.self_attn = EngGPTMoeAttention(config, layer_idx)
+
+        if (layer_idx not in config.mlp_only_layers) and (
+            config.num_experts > 0 and (layer_idx + 1) % config.decoder_sparse_step == 0
+        ):
+            self.mlp = EngGPTMoeSparseMoeBlock(config)
+        else:
+            self.mlp = EngGPTMoeMLP(config, intermediate_size=config.intermediate_size)
+
+        self.input_layernorm = EngGPTMoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.post_attention_layernorm = EngGPTMoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+
+    @deprecate_kwarg("past_key_value", new_name="past_key_values", version="4.58")
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        position_embeddings: tuple[torch.Tensor, torch.Tensor],
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> torch.FloatTensor:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
+            attention_mask (`torch.FloatTensor`, *optional*): attention mask of size
+                `(batch, sequence_length)` where padding elements are indicated by 0.
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            output_router_logits (`bool`, *optional*):
+                Whether or not to return the logits of all the routers. They are useful for computing the router loss,
+                and should not be returned during inference.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+            past_key_values (`Cache`, *optional*): cached past key and value projection states
+            cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
+                Indices depicting the position of the input sequence tokens in the sequence.
+            position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
+                Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
+                with `head_dim` being the embedding dimension of each attention head.
+            kwargs (`dict`, *optional*):
+                Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code
+                into the model
+        """
+        residual = hidden_states
+
+        hidden_states = self.input_layernorm(hidden_states)
+
+        hidden_states, _ = self.self_attn(
+            hidden_states=hidden_states,
+            position_embeddings=position_embeddings,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            cache_position=cache_position,
+            **kwargs,
+        )
+        hidden_states = residual + hidden_states
+
+        residual = hidden_states
+        hidden_states = self.post_attention_layernorm(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+
+        if isinstance(hidden_states, tuple):
+            hidden_states, _ = hidden_states
+        hidden_states = residual + hidden_states
+
+        return hidden_states
+
+
+class EngGPTMoeRotaryEmbedding(nn.Module):
+    inv_freq: torch.Tensor  # fix linting for `register_buffer`
+
+    def __init__(self, config: EngGPTMoeConfig, device=None):
+        super().__init__()
+        # BC: "rope_type" was originally "type"
+        if hasattr(config, "rope_scaling") and isinstance(config.rope_scaling, dict):
+            self.rope_type = config.rope_scaling.get("rope_type", config.rope_scaling.get("type"))
+        else:
+            self.rope_type = "default"
+        self.max_seq_len_cached = config.max_position_embeddings
+        self.original_max_seq_len = config.max_position_embeddings
+
+        self.config = config
+        self.rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type]
+
+        inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device)
+        self.register_buffer("inv_freq", inv_freq, persistent=False)
+        self.original_inv_freq = self.inv_freq
+
+    @torch.no_grad()
+    @dynamic_rope_update  # power user: used with advanced RoPE types (e.g. dynamic rope)
+    def forward(self, x, position_ids):
+        inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1).to(x.device)
+        position_ids_expanded = position_ids[:, None, :].float()
+
+        device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != "mps" else "cpu"
+        with torch.autocast(device_type=device_type, enabled=False):  # Force float32
+            freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2)
+            emb = torch.cat((freqs, freqs), dim=-1)
+            cos = emb.cos() * self.attention_scaling
+            sin = emb.sin() * self.attention_scaling
+
+        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
+
+
+@auto_docstring
+class EngGPTMoePreTrainedModel(PreTrainedModel):
+    config: EngGPTMoeConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["EngGPTMoeDecoderLayer"]
+    _skip_keys_device_placement = ["past_key_values"]
+    _supports_flash_attn = True
+    _supports_sdpa = True
+    _supports_flex_attn = True
+    _can_compile_fullgraph = False  # MoE models don't work with torch.compile (`torch.where(condition)` not supported)
+    _supports_attention_backend = True
+    _can_record_outputs = {
+        "router_logits": OutputRecorder(EngGPTMoeSparseMoeBlock, index=1),
+        "hidden_states": EngGPTMoeDecoderLayer,
+        "attentions": EngGPTMoeAttention,
+    }
+
+
+@auto_docstring
+class EngGPTMoeModel(EngGPTMoePreTrainedModel):
+    def __init__(self, config: EngGPTMoeConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [EngGPTMoeDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
+        )
+        self.norm = EngGPTMoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.rotary_emb = EngGPTMoeRotaryEmbedding(config=config)
+        self.gradient_checkpointing = False
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @check_model_inputs
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        **kwargs: Unpack[TransformersKwargs],
+    ) -> MoeModelOutputWithPast:
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+
+        if use_cache and past_key_values is None:
+            past_key_values = DynamicCache(config=self.config)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embed_tokens(input_ids)
+
+        if cache_position is None:
+            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            cache_position = torch.arange(
+                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+            )
+        if position_ids is None:
+            position_ids = cache_position.unsqueeze(0)
+
+        mask_function = create_causal_mask if self.config.sliding_window is None else create_sliding_window_causal_mask
+        causal_mask = mask_function(
+            config=self.config,
+            input_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+            cache_position=cache_position,
+            past_key_values=past_key_values,
+            position_ids=position_ids,
+        )
+
+        hidden_states = inputs_embeds
+
+        # create position embeddings to be shared across the decoder layers
+        position_embeddings = self.rotary_emb(hidden_states, position_ids)
+
+        for decoder_layer in self.layers[: self.config.num_hidden_layers]:
+            hidden_states = decoder_layer(
+                hidden_states,
+                position_embeddings=position_embeddings,
+                attention_mask=causal_mask,
+                position_ids=position_ids,
+                past_key_values=past_key_values,
+                use_cache=use_cache,
+                cache_position=cache_position,
+                **kwargs,
+            )
+
+        hidden_states = self.norm(hidden_states)
+
+        return MoeModelOutputWithPast(  # only diff with Mistral is the output type, we need MoE
+            last_hidden_state=hidden_states,
+            past_key_values=past_key_values,
+        )
+
+
+def load_balancing_loss_func(
+    gate_logits: Union[torch.Tensor, tuple[torch.Tensor], None],
+    num_experts: Optional[int] = None,
+    top_k=2,
+    attention_mask: Optional[torch.Tensor] = None,
+) -> Union[torch.Tensor, int]:
+    r"""
+    Computes auxiliary load balancing loss as in Switch Transformer - implemented in Pytorch.
+
+    See Switch Transformer (https://huggingface.co/papers/2101.03961) for more details. This function implements the loss
+    function presented in equations (4) - (6) of the paper. It aims at penalizing cases where the routing between
+    experts is too unbalanced.
+
+    Args:
+        gate_logits:
+            Logits from the `gate`, should be a tuple of model.config.num_hidden_layers tensors of
+            shape [batch_size X sequence_length, num_experts].
+        num_experts:
+            Number of experts
+        top_k:
+            The number of experts to route per-token, can be also interpreted as the `top-k` routing
+            parameter.
+        attention_mask (`torch.Tensor`, *optional*):
+            The attention_mask used in forward function
+            shape [batch_size X sequence_length] if not None.
+
+    Returns:
+        The auxiliary loss.
+    """
+    if gate_logits is None or not isinstance(gate_logits, tuple):
+        return 0
+
+    if isinstance(gate_logits, tuple):
+        compute_device = gate_logits[0].device
+        concatenated_gate_logits = torch.cat([layer_gate.to(compute_device) for layer_gate in gate_logits], dim=0)
+
+    routing_weights = torch.nn.functional.softmax(concatenated_gate_logits, dim=-1)
+
+    _, selected_experts = torch.topk(routing_weights, top_k, dim=-1)
+
+    expert_mask = torch.nn.functional.one_hot(selected_experts, num_experts)
+
+    if attention_mask is None:
+        # Compute the percentage of tokens routed to each experts
+        tokens_per_expert = torch.mean(expert_mask.float(), dim=0)
+
+        # Compute the average probability of routing to these experts
+        router_prob_per_expert = torch.mean(routing_weights, dim=0)
+    else:
+        batch_size, sequence_length = attention_mask.shape
+        num_hidden_layers = concatenated_gate_logits.shape[0] // (batch_size * sequence_length)
+
+        # Compute the mask that masks all padding tokens as 0 with the same shape of expert_mask
+        expert_attention_mask = (
+            attention_mask[None, :, :, None, None]
+            .expand((num_hidden_layers, batch_size, sequence_length, top_k, num_experts))
+            .reshape(-1, top_k, num_experts)
+            .to(compute_device)
+        )
+
+        # Compute the percentage of tokens routed to each experts
+        tokens_per_expert = torch.sum(expert_mask.float() * expert_attention_mask, dim=0) / torch.sum(
+            expert_attention_mask, dim=0
+        )
+
+        # Compute the mask that masks all padding tokens as 0 with the same shape of tokens_per_expert
+        router_per_expert_attention_mask = (
+            attention_mask[None, :, :, None]
+            .expand((num_hidden_layers, batch_size, sequence_length, num_experts))
+            .reshape(-1, num_experts)
+            .to(compute_device)
+        )
+
+        # Compute the average probability of routing to these experts
+        router_prob_per_expert = torch.sum(routing_weights * router_per_expert_attention_mask, dim=0) / torch.sum(
+            router_per_expert_attention_mask, dim=0
+        )
+
+    overall_loss = torch.sum(tokens_per_expert * router_prob_per_expert.unsqueeze(0))
+    return overall_loss * num_experts
+
+
+@auto_docstring
+class EngGPTMoeForCausalLM(EngGPTMoePreTrainedModel, GenerationMixin):
+    _tied_weights_keys = ["lm_head.weight"]
+    _tp_plan = {"lm_head": "colwise_rep"}
+    _pp_plan = {"lm_head": (["hidden_states"], ["logits"])}
+
+    def __init__(self, config: EngGPTMoeConfig):
+        super().__init__(config)
+        self.model = EngGPTMoeModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+        self.router_aux_loss_coef = config.router_aux_loss_coef
+        self.num_experts = config.num_experts
+        self.num_experts_per_tok = config.num_experts_per_tok
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_router_logits: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        logits_to_keep: Union[int, torch.Tensor] = 0,
+        **kwargs: Unpack[TransformersKwargs],
+    ) -> MoeCausalLMOutputWithPast:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer, EngGPTMoeForCausalLM
+
+        >>> model = EngGPTMoeForCausalLM.from_pretrained("mistralai/EngGPTMoe-8x7B-v0.1")
+        >>> tokenizer = AutoTokenizer.from_pretrained("mistralai/EngGPTMoe-8x7B-v0.1")
+
+        >>> prompt = "Hey, are you conscious? Can you talk to me?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
+        ```"""
+
+        output_router_logits = (
+            output_router_logits if output_router_logits is not None else self.config.output_router_logits
+        )
+
+        outputs: MoeModelOutputWithPast = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_router_logits=output_router_logits,
+            cache_position=cache_position,
+            **kwargs,
+        )
+
+        hidden_states = outputs.last_hidden_state
+
+        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
+        logits = self.lm_head(hidden_states[:, slice_indices, :])
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits, labels, self.vocab_size, **kwargs)
+
+        aux_loss = None
+        if output_router_logits:
+            aux_loss = load_balancing_loss_func(
+                outputs.router_logits,
+                self.num_experts,
+                self.num_experts_per_tok,
+                attention_mask,
+            )
+            if labels is not None:
+                loss += self.router_aux_loss_coef * aux_loss.to(loss.device)
+
+        return MoeCausalLMOutputWithPast(
+            loss=loss,
+            aux_loss=aux_loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            router_logits=outputs.router_logits,
+        )
+
+
+class EngGPTMoeForSequenceClassification(GenericForSequenceClassification, EngGPTMoePreTrainedModel):
+    pass
+
+
+class EngGPTMoeForTokenClassification(GenericForTokenClassification, EngGPTMoePreTrainedModel):
+    pass
+
+
+class EngGPTMoeForQuestionAnswering(GenericForQuestionAnswering, EngGPTMoePreTrainedModel):
+    base_model_prefix = "transformer"  # For BC, where `transformer` was used instead of `model`
+
+
+__all__ = [
+    "EngGPTMoeForCausalLM",
+    "EngGPTMoeForQuestionAnswering",
+    "EngGPTMoeModel",
+    "EngGPTMoePreTrainedModel",
+    "EngGPTMoeForSequenceClassification",
+    "EngGPTMoeForTokenClassification",
+]

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9f3a9950d12942d3f2c3207b9fd8716c8a0569dadfb828511070b16f5f69886
+size 17080552

tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<unk>"
+}