habryka comments on LLM cognition is probably not human-like

[habryka]

Alas, I can’t find the one or two that I looked at quickly. It came up in a recent Twitter conversation, I think with Quintin?