| { | |
| "_name_or_path": "/data6/models/Meta-Llama-3.1-405B-Instruct", | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 128000, | |
| "eos_token_id": [ | |
| 128001, | |
| 128008, | |
| 128009 | |
| ], | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_size": 16384, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 53248, | |
| "max_position_embeddings": 131072, | |
| "mlp_bias": false, | |
| "model_type": "llama", | |
| "num_attention_heads": 128, | |
| "num_hidden_layers": 126, | |
| "num_key_value_heads": 16, | |
| "pretraining_tp": 1, | |
| "quantization_config": { | |
| "amp": true, | |
| "autoround_version": "0.4.3", | |
| "batch_size": 1, | |
| "bits": 4, | |
| "data_type": "int", | |
| "dataset": "NeelNanda/pile-10k", | |
| "enable_minmax_tuning": true, | |
| "enable_norm_bias_tuning": false, | |
| "enable_quanted_input": true, | |
| "gradient_accumulate_steps": 4, | |
| "group_size": -1, | |
| "iters": 200, | |
| "low_gpu_mem_usage": true, | |
| "lr": 0.005, | |
| "minmax_lr": 0.005, | |
| "nsamples": 128, | |
| "quant_method": "auto-round", | |
| "scale_dtype": "torch.float16", | |
| "seqlen": 2048, | |
| "sym": true, | |
| "to_quant_block_names": null | |
| }, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 500000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.47.0", | |
| "use_cache": true, | |
| "vocab_size": 128256 | |
| } | |