OpenSourceRonin's picture
Upload model Meta-Llama-3.1-405B-Instruct-v16-k32768-32768-woft
cef8595 verified
raw
history blame contribute delete
338 Bytes
{
"attn_implementation": "flash_attention_2",
"bos_token_id": 128000,
"do_sample": true,
"eos_token_id": [
128001,
128008,
128009
],
"num_assistant_tokens": 5,
"num_assistant_tokens_schedule": "heuristic",
"return_legacy_cache": true,
"temperature": 0.6,
"top_p": 0.9,
"transformers_version": "4.45.1"
}