| { | |
| "_name_or_path": "checkpoints-preferences/preferences_princeton-nlp--Sheared-LLaMA-1.3b_bsz512_lr5e-5_epochs2_warmup0.1_conf0.5_labeltemp1.0_datasetnew_labelall_main/", | |
| "architectures": [ | |
| "LlamaForSequenceClassification" | |
| ], | |
| "attention_bias": false, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 2048, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1", | |
| "2": "LABEL_2", | |
| "3": "LABEL_3" | |
| }, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 5504, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1, | |
| "LABEL_2": 2, | |
| "LABEL_3": 3 | |
| }, | |
| "max_position_embeddings": 4096, | |
| "model_type": "llama", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 24, | |
| "num_key_value_heads": 16, | |
| "pad_token_id": 0, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.35.0", | |
| "use_cache": true, | |
| "vocab_size": 32000 | |
| } | |