On: How do neural networks learn?

[swalsh]

I've been trying to figure out how LLMs learn logic/reasoning. It's just not intuitive to me how that works.

[mikewarot → swalsh]

It's the same way we do it, form a number of possible variants and use the ones that work best.

They have the advantage of millions or even billions of times more compute to throw at the learning process. Something that might be a one in a million insight happens consistently at that scale.