3 ms·
I’ve found that the RLHF’d ChatGPT is way too submissive these days. I really do not enjoy asking for minor clarification and getting back “I apologize for the
by _nhynes 3y ago
I’ve found that the RLHF’d ChatGPT is way too submissive these days. I really do not enjoy asking for minor clarification and getting back “I apologize for the confusion…” followed by a completely and incorrectly revised reply.
- minimaxir 3y agoUsing the raw API with some system prompt engineering gets ChatGPT to behave very well, even deprogramming the RLHF a bit.