Humans also respond differently when prompted in different ways. For example, politeness often begets politeness. I would expect that to be reflected in training data.

▲

mlsu 2 hours ago | parent | next [-]

If I, a moron, hire a PhD to crack a tough problem for me, I don't need to go back and forth prompting him at a PhD level. I can set him loose on my problem and he'll come back to me with a solution.

	▲	nl 44 minutes ago \| parent \| next [-]
		> hire a PhD to crack a tough problem for me, I don't need to go back and forth prompting him at a PhD level. I can set him loose on my problem and he'll come back to me with a solution. In my experience with many PhDs they are just as prone to getting off track or using their pet techniques as LLMs! And many find it very hard to translate their work into everyday language too...
	▲	Herring 2 hours ago \| parent \| prev [-]
		Well if it ever gets to be a full replacement for phds, you’ll know cause it will have already replaced you.

▲

HPsquared 2 hours ago | parent | prev [-]

I think that's what is happening. It's simulating a conversation, after all. A bit like code switching.