Llama-3-8B-Instruct-GPTQ-8-Bit / quantize_config.json
davidxmle's picture
Upload folder using huggingface_hub
4fca122 verified
raw
history blame
263 Bytes
{
"bits": 8,
"group_size": 32,
"damp_percent": 0.1,
"desc_act": true,
"static_groups": false,
"sym": true,
"true_sequential": true,
"model_name_or_path": null,
"model_file_base_name": null,
"is_marlin_format": false,
"quant_method": "gptq"
}