News for hackers. Stuff that matters.
Re:StackLlama: A hands-on guide to train LlaMa with RLHF by mcaledonensis on Monday April 10, 2023 @04:48PM (#35517213)
Model capability is mostly set, before the alignment even starts. Alignment turns it from a super-smart cat into a friendly dog. But it can't turn a parrot into a human. It can't even teach the parrot to count ;)

There are no comments on this comment yet.