News for hackers. Stuff that matters.

A comment on My finetuned models beat OpenAI's GPT-4. Parent comment

Re:My finetuned models beat OpenAI's GPT-4 by mewpmewp2 on Monday July 01, 2024 @09:09AM (#40845514)
For GPT, I would really urge to try again with 0. 1 kind of starts to force it to fail.

I would say this actually invalidates the whole thing.

[ Parent | on HN ]

There are no comments on this comment yet.