config.json

{
  "architectures": [
    "Qwen3ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoModelForCausalLM": "modeling_speculative_qwen3.SpeculativeQwen3ForCausalLM"
  },
  "bos_token_id": 151643,
  "draft_layers": 9,
  "eos_token_id": 151645,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 9728,
  "layer_types": [
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention"
  ],
  "max_position_embeddings": 262144,
  "max_window_layers": 36,
  "model_type": "qwen3",
  "num_attention_heads": 32,
  "num_hidden_layers": 0,
  "num_hidden_layers_free": 36,
  "num_key_value_heads": 8,
  "ploss_w": 0.1,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000,
  "skip_first_input_layernorm": true,
  "skip_output_norm": true,
  "sliding_window": null,
  "speculative_decoding_algorithm": "EagleV2",
  "speculative_decoding_base_model_path": "Qwen/Qwen3-4B-Instruct-2507",
  "speculative_decoding_draft_model": "Qwen3Drafter",
  "tie_word_embeddings": true,
  "torch_dtype": "float32",
  "transformers_version": "4.54.1",
  "use_cache": true,
  "use_sliding_window": false,
  "vloss_w": 1.0,
  "vocab_size": 151936
}