Yuan2-M32-hf-int8 / config.json
IEIT-Yuan's picture
hugging
e643fa3
{
"_from_model_config": true,
"_name_or_path": "/temp_data/LLM_test/MOE/Yuan2-M32",
"architectures": [
"YuanForCausalLM"
],
"attention_projection_size": 4096,
"auto_map": {
"AutoConfig": "configuration_yuan.YuanConfig",
"AutoModelForCausalLM": "yuan_hf_model.YuanForCausalLM"
},
"bos_token_id": 77185,
"causal_mask": true,
"dropout": 0,
"eod_token": 77185,
"eod_token_id": 77185,
"eos_token_id": 77185,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 8192,
"mask_token_id": 77185,
"max_position_embeddings": 4096,
"model_max_length": 8192,
"model_type": "yuan",
"moe_config": {
"ffn_hidden_size": 8192,
"gated_linear_unit": true,
"moe_num_experts": 32,
"moe_top_k": 2,
"norm_topk_prob": true
},
"num_attention_heads": 16,
"num_hidden_layers": 24,
"output_router_logits": true,
"pad_token_id": 77185,
"quantization_config": {
"bits": 8,
"checkpoint_format": "gptq",
"damp_percent": 0.01,
"desc_act": false,
"group_size": 128,
"model_file_base_name": "gptq_model-8bit-128g",
"model_name_or_path": "/temp_data/LLM_test/MOE/Yuan2-M32-GPTQ-int8",
"quant_method": "gptq",
"static_groups": false,
"sym": true,
"true_sequential": true
},
"reset_attention_mask": false,
"reset_position_ids": true,
"rms_norm_eps": 1e-06,
"sep_token": 77187,
"sep_token_id": 77185,
"tokenizer_class": "YuanTokenizer",
"torch_dtype": "bfloat16",
"transformers_version": "4.39.3",
"use_cache": true,
"use_flash_attention": true,
"use_loss_mask": false,
"use_moe": true,
"vocab_size": 135040
}