Llama3.2-1B-R1-AKD / generation_config.json
ilyasoulk's picture
Add akd for deepseek distill r1 qwen and llama 3.2 1b
8e022e4
raw
history blame contribute delete
184 Bytes
{
"bos_token_id": 128000,
"do_sample": true,
"eos_token_id": [
128001,
128008,
128009
],
"temperature": 0.6,
"top_p": 0.9,
"transformers_version": "4.46.1"
}