@nosilverv this isn't unusual if you realise that a LLM is basically a wrapper around the compression of all the text in a corpus, to compress gud it has to find some way of reducing the complexity.…

Posters: pachabelcanon. pachabelcanon

2 likes recorded by the source corpus.

Topics: Epistemics

View Post: https://x.com/pachabelcanon/status/1762450751253623292