{
  "_name_or_path": "/home/ry21/Generalizable-Reward-Model/reward_models/reward_models_new/gemma-2-2b-it_GRM_gemma2-2B_mixdata_length3000_mlp_sftreg0.01_fulltrain_lr5e-6_5e-06/logs/checkpoint-1072_0.8119",
  "architectures": [
    "Gemma2ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attn_logit_softcapping": 50.0,
  "auto_map": {
    "AutoModel": "model.AutoModelForCausalLMWithValueHead",
    "AutoModelForSequenceClassification": "model.AutoModelForCausalLMWithValueHead"
  },
  "bos_token_id": 2,
  "cache_implementation": "hybrid",
  "eos_token_id": [
    1,
    107
  ],
  "final_logit_softcapping": 30.0,
  "head_dim": 256,
  "hidden_act": "gelu_pytorch_tanh",
  "hidden_activation": "gelu_pytorch_tanh",
  "hidden_size": 2304,
  "initializer_range": 0.02,
  "intermediate_size": 9216,
  "max_position_embeddings": 8192,
  "model_type": "gemma2",
  "num_attention_heads": 8,
  "num_hidden_layers": 26,
  "num_key_value_heads": 4,
  "pad_token_id": 1,
  "query_pre_attn_scalar": 256,
  "rms_norm_eps": 1e-06,
  "rope_theta": 10000.0,
  "sliding_window": 4096,
  "torch_dtype": "float16",
  "transformers_version": "4.44.2",
  "use_cache": true,
  "vhead_layer_type": "mlp",
  "vhead_num_layers": 1,
  "vhead_num_neurons": 1024,
  "vocab_size": 256000
}