DachengLi1 / LongChat

Official repository for LongChat and LongEval

Longchat inference configuration

SeekWrldTea opened this issue a year ago · comments

TheDevDigger commented a year ago

What do you guys recommend as for the inference configuration for : temperature and repeat penalty etc ... ?

Dacheng Li commented a year ago

@SeekWrldTea temp=1.0, penalty=1.0: https://github.com/lm-sys/FastChat/blob/main/fastchat/serve/inference.py#L57