qwopus3.5-9b-coder-awq-int4 / generation_config.json
compute1's picture
Add INT4 quantized Qwopus3.5-9B-Coder with linear-attn bf16 patch
3d2c9cf verified
Raw
History Blame Contribute Delete
141 Bytes
{
"_from_model_config": true,
"eos_token_id": 248046,
"pad_token_id": 248055,
"transformers_version": "5.9.0",
"use_cache": true
}