3 ms·
How can i locate this study?. I think you are misrepresenting something. In the gpt4 paper they specifically address this, and find that "Averaged across all e
by mov_eax_ecx 3y ago
How can i locate this study?. I think you are misrepresenting something.
In the gpt4 paper they specifically address this, and find that "Averaged across all exams, the base model achieves a score of 73.7% while the RLHF model achieves a score of 74.0%, suggesting that post-training does not substantially alter base model capability."
- nicce 3y agoThe problem with these studies is that we really still don’t know. Nobody can replicate the papers of OpenAI.
- adamsb6 3y agoGiven the homogeneity of responses on taboo subjects, there's probably something exogenous to the model at work.
- galaxytachyon 3y agoFound it, it is a pretty recent paper. https://arxiv.org/pdf/2308.13449.pdf https://arxiv.org/pdf/2308.13449.pdf