{
  "_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
  "architectures": ["QwenLMHeadModel"],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 5504,
  "max_position_embeddings": 8192,
  "model_type": "qwen2",
  "num_attention_heads": 16,
  "num_hidden_layers": 24,
  "rms_norm_eps": 1e-05,
  "rope_theta": 10000.0,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.39.3",
  "use_cache": true,
  "vocab_size": 151936
}