CIRCLE WITH A DOT

@mcc @glyph Right, which is why they are called "model weights" and not "model coin flips." Models are non-deterministic at the token level but pretty darn consistent at the macro level, which is why ChatGPT articles are so easy to spot. "It's not X, it's Y"; numbered lists; boldface, etc.

@mcc @glyph I don't think alignment has anything to do with determinism. People are non-deterministic but a person can absolutely be ethnically aligned (or not).