Slacker News Slacker News logo featuring a lazy sloth with a folded newspaper hat
  • top
  • new
  • show
  • ask
  • jobs
Library
← Back to context

Comment by heavyset_go

1 month ago

The HF part of RLHF to refine the output of LLMs also happens in these places

1 comment

heavyset_go

Reply

astrange  1 month ago

Note RLHF can only perform selection on existing model outputs, adding new data is SFT or else just more pretraining.

ChatGPT speaking African English was mostly just 3.5. 4o speaks like a TikTok user from LA. 5 seems kind of generic.

Slacker News

Product

  • API Reference
  • Hacker News RSS
  • Source on GitHub

Community

  • Support Ukraine
  • Equal Justice Initiative
  • GiveWell Charities