Meta-Llama-3-8B-Instruct-GPTQ-Int8/quantize_config.json

{
  "bits": 8,
  "dataset": "sharegpt-gpt4-mini",
  "group_size": 128,
  "damp_percent": 0.1,
  "desc_act": false,
  "sym": true,
  "true_sequential": true,
  "quant_method": "gptq",
  "modules_in_block_to_quantize": null
}