News for hackers. Stuff that matters.
Re:Stability AI releases StableVicuna, a RLHF LLM Chatbot by smusamashah on Friday April 28, 2023 @03:10PM (#35745740)
Why is it trained on GPT generated content? Won't it reduced the content quality instead of improving it? GPT generates lots of fluff which will only pollute the new models further.
Re:Stability AI releases StableVicuna, a RLHF LLM Chatbot by nickthegreek on Friday April 28, 2023 @03:24PM (#35745954)
The vicuna model has shown that there is value in this method.

https://vicuna.lmsys.org/