Llama-3-8B-Instruct-v0.5-AWQ / quant_config.json
Suparious's picture
adding quant config
49f4aed verified
raw
history blame
No virus
82 Bytes
{
"zero_point": true,
"q_group_size": 128,
"w_bit": 4,
"version": "GEMM"
}