Personas are just a form of generalization
kind: experiment
track:Highlight text on any card (body, plan, or an event) to anchor a comment, or leave a whole-task comment here. Mention @claude to summon a reply.
No comments yet.
Scaling laws of generalization Look into theories of generalization Large models learn more spurious correlations and bad generalization -> Because they are generalizing better Unless you interleave stuff specifically We know larger models are better at generalizing -> but this can be bad Jason -> no correlation between model size and EM amount, but this was confounded
Activity