how can i limit the generate length , i haven't see the new max length in chat().
#6
by 520jefferson - opened
i want to know how to control the lenght when i use the model to chat .
I see max_length: int = 8192 in chat() guess that is your new_max_length
In our latest github code, the code is unified into max_new_tokens
ZHANGYUXUAN-zR changed discussion status to closed