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

Well, I'm thinking I must be missing some nuance, it seems like there must be some session state stored in the LLM somehow, otherwise it would have to re-parse the entire chat session from scratch to generate each succeeding token. But maybe it does do that. I'm trying to understand if it is learning, or building a model of the world in any persistent way how that is happening and where it exists, if at all.


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

Search: