sft-conta-deepseek-distill-llama3-8b / generation_config.json
ko120's picture
Initial upload
44026fa verified
raw
history blame contribute delete
181 Bytes
{
"_from_model_config": true,
"bos_token_id": 128000,
"do_sample": true,
"eos_token_id": 128001,
"temperature": 0.6,
"top_p": 0.95,
"transformers_version": "4.52.4"
}