Comment by ModernMech
13 hours ago
I always thought it could be because volume-wise, most English prose is probably marketing copy and actual clickbait; so when you train on the entire Internet, you get a troll adept at writing ads. Then people ask AdBot2000 to write a novel and are upset it reads like the next iPhone launch site.
Nah, I think this is a common misunderstanding of how LLMs work, where people think that they mimic the pre-training data. Stylistically everything you see is an artifact of post-training, which is from reinforcement learning not from absorbing mass amounts of text. At some point a person or more recently a bot gave a thumbs up to an A/B tested response including em-dashes and claudisms galore.
> Stylistically everything you see is an artifact of post-training,
It is still not exactly clear if it is true or not. Unless we have base "pt" snaphot of Claude we can't say one way or another. I've played a bit with base models of Nemo, Gemma etc and they all had tics, not much different from RLHFed instruct versions.
Yes. This completely explains sycophancy at least.
So question then, why is it so hard to make an ai that doesn’t do these things? And why do Claude and ChatGPT have the same -isms? They’re both doing the same a/b post training with the same decisions?
It would require changing humans first.
You don't blame the puddle for taking the shape of the hole.
There's layers, some of token selection is fingerprinting https://github.com/google-deepmind/synthid-text
Yeah, but I understand that fingerprinting is essentially a pseudorandom overlay onto a pseudorandom base signal. And unless you have access to both the random number generators and the weights, I don't think you can detect it?
So "fingerprinting" operates on a totally different and basically invisible level, as opposed to the obvious stylistic patterns that the average programmer can identify in about 2 sentences.
No, there's no reason chatbot behavior would have anything to do with frequency of text in pretraining.