Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The model isn't being trained as time goes on.


The RLHF layer is, and that layer is important.

That it does any of this is the specific reason for the story you're commenting on, and why putting data into it isn't like putting the same data into e.g. a Google Docs spreadsheet.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: