quantize_config.json

{
  "bits": 8,
  "group_size": 128,
  "damp_percent": 0.01,
  "desc_act": false,
  "static_groups": false,
  "sym": true,
  "true_sequential": true,
  "model_name_or_path": "/temp_data/LLM_test/MOE/Yuan2-M32-GPTQ-int8",
  "model_file_base_name": "gptq_model-8bit-128g",
  "quant_method": "gptq",
  "checkpoint_format": "gptq"
}