初始化项目,由ModelHub XC社区提供模型
Model: astronomer/Llama-3-8B-Instruct-GPTQ-4-Bit Source: Original Platform
This commit is contained in:
13
quantize_config.json
Normal file
13
quantize_config.json
Normal file
@@ -0,0 +1,13 @@
|
||||
{
|
||||
"bits": 4,
|
||||
"group_size": 128,
|
||||
"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"
|
||||
}
|
||||
Reference in New Issue
Block a user