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

The training code, parameters and training dataset got released, and the training process only takes a few hours. Anyone can recreate it.


Actually it seems that Alpaca is not actually released. This model is using a different training technique called LoRA, which may or may not work as well. People are making a lot of unsubstantiated performance claims, saying this is as good as GPT-3 when not nearly enough evaluation has been done to demonstrate that yet.




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

Search: