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

Definitely has trended upward, there’s no special number. It’s just a matter of how much compute, storage, time to allocate to that part of the architecture.


Has it? It's my understanding that GPT-3 was 12,288 and GPT-4 went down to 3072.


I’m not aware that internal embedding dimension size has been made public for Gpt4 et al.

In general for models we know its trended upward, but for sure it’d be interesting to know what they’re using now.




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

Search: