ColossalChat tends to defend on its FAKE knowledge and still be authoritative. #3336
pps43
started this conversation in
Community | General
Replies: 1 comment
-
|
The current dataset is quite limited, so we can't expect it to perform as well as ChatGPT. As releaved by OpenAI, they spent months on the alignment, which is insufficient on our ColossalChat. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
The right side is from ChatGPT. I don't quite understand why they give totally different answers on a technique fact (ChatGPT is right because I know PhysX library). I upload the chat history to see if it helps to tune this model better.
Beta Was this translation helpful? Give feedback.
All reactions