TinyLlama-1.1B-Chat-v1.0-AWQ / quant_config.json
TheBloke's picture
AWQ model commit
f68464f
raw
history blame
126 Bytes
{
"zero_point": true,
"q_group_size": 128,
"w_bit": 4,
"version": "GEMM",
"modules_to_not_convert": null
}