They're not 3.5 models, which are 175B param, they're 220B and the idea is the f...

		brandall10 on July 19, 2023 \| parent \| context \| favorite \| on: GPT-4 is getting worse over time, not better They're not 3.5 models, which are 175B param, they're 220B and the idea is the fine tuning is different for each which is where the 'expert' comes in.

Ah, that makes sense, thanks for clarifying