I liked reading this article. Its cool to really poke into the hidden perplexity behind patterns of ‘thought’ in llms. They aren’t merely simple ‘auto complete’.
The finding that claude does math in a different way then it says it does and can anticipate words ahead of generation time are facinating.
This is insane it does seem to work like a brain in a way, especially in that we don’t understand either of them
Maybe this leads to further understanding of neuroscience some how?