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

This particular model is super bad at 600 deferent tasks. At it's size you'd expect it to be mediocre at best at even one of them, so it's still very impressive. Fascinating research, can't wait to see if it's generalizing and how, not sure how overall significant it is


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

Search: