config.json

{
  "_name_or_path": "/mnt/petrelfs/wuhaoyuan.d/models/Qwen1.5-14B",
  "adapter_dim": 512,
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_qwen2idae.Qwen2idaeConfig",
    "AutoModelForCausalLM": "modeling_qwen2idae.Qwen2ForCausalLM"
  },
  "bos_token_id": 151643,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 13696,
  "max_position_embeddings": 32768,
  "max_window_layers": 35,
  "model_type": "qwen2idae",
  "moe_dtype": "bfloat16",
  "moe_scaling": 0.25,
  "num_attention_heads": 40,
  "num_experts": 16,
  "num_hidden_layers": 40,
  "num_key_value_heads": 40,
  "output_router_logits": false,
  "rms_norm_eps": 1e-06,
  "rope_theta": 1000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "topk": 4,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.37.2",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 152064
}