{
  "_name_or_path": "/root/Models/LongBigPhi91",
  "architectures": [
    "Phi3ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "embd_pdrop": 0.0,
  "eos_token_id": 32000,
  "hidden_act": "silu",
  "hidden_actt": "silu",
  "hidden_size": 4608,
  "initializer_range": 0.02,
  "intermediate_size": 8192,
  "max_position_embeddings": 131072,
  "model_type": "phi3",
  "norm_divisor": null,
  "num_attention_heads": 48,
  "num_hidden_layers": 48,
  "num_key_value_heads": 48,
  "original_max_position_embeddings": 131072,
  "pad_token_id": 32000,
  "resid_pdrop": 0.0,
  "return_states_at_layer": null,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "sliding_window": 262144,
  "start_at_layer": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.0",
  "use_cache": false,
  "vocab_size": 32064
}