AMA on the 17th of December with OpenAI's API Team: Post Your Questions Here

can preference tuning be used to encourage longer answers (say 10+ pages) and better use of output context ?

1 Like