llama-2-ko-7b / generation_config.json
beomi's picture
Add 5B korean trained model, test version.
d8eaedb
raw
history blame
159 Bytes
{
"_from_model_config": true,
"bos_token_id": 1,
"eos_token_id": 2,
"max_length": 4096,
"pad_token_id": 0,
"transformers_version": "4.28.0.dev0"
}