alt Hacker NewsI found the article interesting, humans are flawed enough in a variety of aspects that discovering a convergence here is noteworthy. I don't take it as a sign the LLMs believe anything in particular, they are not capable of belief, as you said.
Everything that comes from an LLM is a human expression, on some level, but if you asked a sampling of humans the same question I wonder if it would come out the same way. How would you explain the difference? What makes that answer more probable?
Replies
> Everything that comes from an LLM is a human expression, on some level, but if you asked a sampling of humans the same question I wonder if it would come out the same way. How would you explain the difference? What makes that answer more probable?
RLHF has a way of squashing variance. If the different models are trained on each other or on the same human feedback, then the training would reduce the variety of answers while the similarity of feedback would lead the common answer to be similar.
The different models may prefer the same controversial takes for the same reason that they tend to prefer the same jokes and names. Maybe the authors of the article would have had better luck if they'd used verbalized sampling prompts or something like it.[1]
I agree that LLMs are still just a way to sample from a probabilistic model of things humans say. And in this case the model seems to be pretty darn accurate.
A pretty staggering amount of human discourse consists of ad hominem attacks and accusing each other of being foolishly committed to various real or perceived allegiances. For example the current version of Godwin’s Law around here has conversations gravitating toward people accusing each other of being AI boosters or AI haters.
> discovering a convergence here is noteworthy
The LLMs are all trained on the same broad corpus, which includes untold quantities of output from every other LLM, not to mention that companies are distilling their rivals left and right. It would be more interesting if they didn't converge.
> if you asked a sampling of humans the same question I wonder if it would come out the same way. How would you explain the difference?
If you asked a sampling of humans what an AI's most controversial opinion about us is it would be very similar. Or, to be more specific, if you asked a sampling of science fiction writers that question it would be very similar, because that is what the AI training data is based on.
The people who wrote a lot about a topic is who the AI mimics. So the people who wrote a lot about AI is therefore the people who the AI mimics when asked about AI.