bertin-gpt-j-6B-alpaca-8bit-128g / quantize_config.json
RedXeol's picture
Upload 13 files
b30a639
raw
history blame
123 Bytes
{
"bits": 8,
"group_size": 128,
"damp_percent": 0.01,
"desc_act": false,
"sym": true,
"true_sequential": true
}