lfm-Nanotron / W2A16-AutoRound /generation_config.json
Ex0bit's picture
Add W2A16 AutoRound quantized: generation_config.json
a4bb9a5 verified
{
"_from_model_config": true,
"bos_token_id": 1,
"do_sample": true,
"eos_token_id": 7,
"min_p": 0.15,
"pad_token_id": 0,
"temperature": 0.3,
"transformers_version": "4.57.6"
}