Comment on My finetuned models beat OpenAI's GPT-4Comments−XiphiasX2y1) beat at what? 2) do they beat Claude 3.5 Sonnet?−input_sh2yHave you tried clicking on the link and finding out?−singularity20012yJust in the task of structured data extractionSo very misleading title−furyofantares2ySo very misleading titleEh, I can see that, but to me "finetuned model" pretty strongly implies some specific task−freehorse2yDid you read the article or just the title? It is all explained there.
Comments
1) beat at what? 2) do they beat Claude 3.5 Sonnet?
Have you tried clicking on the link and finding out?
Just in the task of structured data extraction
So very misleading title
Eh, I can see that, but to me "finetuned model" pretty strongly implies some specific task
Did you read the article or just the title? It is all explained there.