On: Learning to Reason with LLMs

[spoonfeeder006]

So how is the internal chain of thought represented anyhow? What does it look like when someone sees it?