open-llava-next-llama3-8b / generation_config.json
chenlin
init
d9717b6
raw
history blame contribute delete
240 Bytes
{
"attn_implementation": "flash_attention_2",
"bos_token_id": 128000,
"do_sample": true,
"eos_token_id": [
128001,
128009
],
"max_length": 4096,
"temperature": 0.6,
"top_p": 0.9,
"transformers_version": "4.37.2"
}