HFXM/DynamicRules_ScoreRM
03
1{2 "_name_or_path": "meta-llama/Meta-Llama-3-8B",3 "architectures": [4 "LlamaForSequenceClassification"5 ],6 "attention_bias": false,7 "attention_dropout": 0.0,8 "bos_token_id": 128000,9 "eos_token_id": 128001,10 "hidden_act": "silu",11 "hidden_size": 4096,12 "id2label": {13 "0": "LABEL_0"14 },15 "initializer_range": 0.02,16 "intermediate_size": 14336,17 "label2id": {18 "LABEL_0": 019 },20 "max_position_embeddings": 8192,21 "mlp_bias": false,22 "model_type": "llama",23 "num_attention_heads": 32,24 "num_hidden_layers": 32,25 "num_key_value_heads": 8,26 "pad_token_id": 128001,27 "pretraining_tp": 1,28 "problem_type": "regression",29 "rms_norm_eps": 1e-05,30 "rope_scaling": null,31 "rope_theta": 500000.0,32 "tie_word_embeddings": false,33 "torch_dtype": "bfloat16",34 "transformers_version": "4.43.4",35 "use_cache": true,36 "vocab_size": 12825637}38 