Text Generation
Transformers
Safetensors
English
llama
causal-lm
text-generation-inference
4-bit precision
gptq
stable-vicuna-13B-GPTQ / quantize_config.json
TheBloke's picture
Create quantize_config.json
ecf7c95
raw
history blame
81 Bytes
{
"bits": 4,
"damp_percent": 0.01,
"desc_act": false,
"group_size": 128
}