Text Generation
Transformers
Safetensors
English
stablelm_epoch
causal-lm
conversational
custom_code
4-bit precision
gptq
stablelm-zephyr-3b-GPTQ / quantize_config.json
TheBloke's picture
GPTQ model commit
343794e
{
"bits": [
4
],
"group_size": [
32
],
"damp_percent": [
0.1
],
"desc_act": [
true
],
"sym": true,
"true_sequential": true
}