Inlay

New blogpost: We Don't Understand Neural Networks At The Algorithmic Level kripken.github.io/blog/neurosc... So, why is this even a question? Don't scientists agree on whether we understand LLMs?

1d

We Don't Understand Neural Networks At The Algorithmic Level

The largest ongoing debate about AI is “Are Large Language Models (LLMs) intelligent?” That makes sense, at least: the evidence is ambiguous and the stakes a...

kripken.github.io

A. H. Zakai

This preprint makes the "we *do* understand LLMs" case better than I've ever seen it. But, while they are right on many things, a lot remains that we don't understand. My post tries to explain what exactly is missing and why their arguments don't address it. Here are my three main points:

1d