{ | |
"_name_or_path": "/mnt/f/git_AIResearch/!models_quantization/Yi-34B-200K", | |
"architectures": [ | |
"YiForCausalLM" | |
], | |
"auto_map": { | |
"AutoConfig": "configuration_yi.YiConfig", | |
"AutoModel": "modeling_yi.YiModel", | |
"AutoModelForCausalLM": "modeling_yi.YiForCausalLM" | |
}, | |
"bos_token_id": 1, | |
"eos_token_id": 2, | |
"hidden_act": "silu", | |
"hidden_size": 7168, | |
"initializer_range": 0.02, | |
"intermediate_size": 20480, | |
"max_position_embeddings": 200000, | |
"model_type": "Yi", | |
"num_attention_heads": 56, | |
"num_hidden_layers": 60, | |
"num_key_value_heads": 8, | |
"pad_token_id": 0, | |
"quantization_config": { | |
"bits": 4, | |
"group_size": 128, | |
"quant_method": "awq", | |
"version": "gemm", | |
"zero_point": true | |
}, | |
"rms_norm_eps": 1e-05, | |
"rope_theta": 5000000.0, | |
"tie_word_embeddings": false, | |
"torch_dtype": "float16", | |
"transformers_version": "4.35.0", | |
"use_cache": true, | |
"vocab_size": 64000, | |
"quantization_config": { | |
"quant_method": "awq", | |
"zero_point": true, | |
"group_size": 128, | |
"bits": 4, | |
"version": "gemm" | |
} | |
} | |