{
  "_name_or_path": "./checkpoints/TrafficLLM_3dataset_wd_llama3_pems08-1/full_model",
  "architectures": [
    "STLlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 128000,
  "eos_token_id": 128001,
  "freeze_st_mlp_adapter": false,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 14336,
  "lin_hidden_size": 128,
  "max_position_embeddings": 131072,
  "mlp_bias": false,
  "model_type": "STLlama",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 8,
  "pretrain_ST_model_path": "./pretrain_encoder/experiments/localgat_alldataset/alldataset/final_model_s2023.pt",
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "factor": 8.0,
    "type": "linear"
  },
  "rope_theta": 500000.0,
  "sep_st_conv_front": false,
  "st_hidden_size": 64,
  "st_select_layer": -2,
  "st_tower": "ST_Encoder",
  "tie_word_embeddings": false,
  "time_steps": 12,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.41.1",
  "tune_st_mlp_adapter": true,
  "use_cache": false,
  "use_st_proj": true,
  "use_st_start_end": true,
  "vocab_size": 128260
}