quantize_config.json

{
  "bits": 4,
  "group_size": 128,
  "sym": true,
  "data_type": "int",
  "nsamples": 512,
  "autoround_version": "0.5.1",
  "quant_method": "gptq",
  "desc_act": false,
  "true_sequential": false,
  "damp_percent": 0.01
}