config.json

{
  "architectures": [
    "DebertaV2Model"
  ],
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 1024,
  "initializer_range": 0.02,
  "intermediate_size": 4096,
  "layer_norm_eps": 1e-07,
  "legacy": true,
  "max_position_embeddings": 512,
  "max_relative_positions": -1,
  "model_type": "deberta-v2",
  "norm_rel_ebd": "layer_norm",
  "num_attention_heads": 16,
  "num_hidden_layers": 24,
  "pad_token_id": 0,
  "pooler_dropout": 0,
  "pooler_hidden_act": "gelu",
  "pooler_hidden_size": 1024,
  "pos_att_type": [
    "p2c",
    "c2p"
  ],
  "position_biased_input": false,
  "position_buckets": 256,
  "relative_attention": true,
  "share_att_key": true,
  "transformers_version": "4.57.6",
  "type_vocab_size": 0,
  "vocab_size": 128100,
  "open_jev": {
    "description": "Typed-decision scorer: one state plus typed questions in one sequence; logits per (question, option) pair from span pooling.",
    "markers": [
      "[STATE]",
      "[Q]",
      "[OPT]"
    ],
    "max_state_tokens": 256,
    "max_len": 512,
    "temperature": 1.05,
    "pool": "span",
    "inputs": "input_ids, attention_mask, seg (span slot per token, -1 elsewhere), pair_q, pair_opt (slot ids per pair)",
    "output": "logits (batch, pairs); softmax within each question's pairs after dividing by temperature"
  },
  "transformers.js_config": {
    "use_external_data_format": {
      "model.onnx": 2,
      "model_fp16.onnx": 1,
      "model_q4.onnx": 1,
      "model_q4f16.onnx": 1
    },
    "dtype": "fp16"
  }
}