RogerDearnaley comments on Worries about latent reasoning in LLMs

[RogerDearnaley]

Well summarized — very similar to the conclusions I’d previously reached when I read the paper.