adjohn1313/wizard_sft_explainable_rlhf_6k

1{2  "_name_or_path": "/home/users/aj391/code/explainable_trl/explainable-merged_wizard/",3  "architectures": [4    "LlamaForCausalLM"5  ],6  "attention_bias": false,7  "attention_dropout": 0.0,8  "bos_token_id": 1,9  "eos_token_id": 2,10  "hidden_act": "silu",11  "hidden_size": 5120,12  "initializer_range": 0.02,13  "intermediate_size": 13824,14  "max_position_embeddings": 2048,15  "max_sequence_length": 2048,16  "model_type": "llama",17  "num_attention_heads": 40,18  "num_hidden_layers": 40,19  "num_key_value_heads": 40,20  "pad_token_id": 0,21  "pretraining_tp": 1,22  "quantization_config": {23    "backend": "autoawq",24    "bits": 4,25    "do_fuse": false,26    "fuse_max_seq_len": null,27    "group_size": 128,28    "modules_to_fuse": null,29    "modules_to_not_convert": null,30    "quant_method": "awq",31    "version": "gemm",32    "zero_point": true33  },34  "rms_norm_eps": 1e-06,35  "rope_scaling": null,36  "rope_theta": 10000.0,37  "tie_word_embeddings": false,38  "torch_dtype": "float16",39  "transformers_version": "4.38.1",40  "use_cache": true,41  "vocab_size": 3200042}43