Re:Major Outage Across ChatGPT and API
by nomel on Wednesday November 08, 2023 @11:28PM
(#38201192)
I subscribe to the belief that, for a chat model with the same parameters, creativity will be proportional to tendency to hallucinate, and inversely proportional to the factual answers. I suspect an unaligned model, without RLHF, wouldn't adhere to this.
[ Parent
| on HN ]
There are no comments on this comment yet.