Link to originalMany naive things become possible in/with CLIP’s/LLM representation space, such as
- specifying goal states in natural language
- stopping simulation/training once change in latent space plateaus (magnitude of clip vector change)
- (heuristic metrics that fall short in other cases)
Link to original
- same in biology, evolution works very different with a competent substrate (cells that can already solve problems)!!
Link to originalOpen ended discovery leads to fundamentally different organization/ representation: Representations are more meaningful / composable / efficient / natural
Explains DNA to some extent: Perturbation is sensible / canalized.
Should representations look like DNA?
If you “think open-endedly” (not optimizing for one specific (end-)goal), do you obtain representations that are different than those who don’t? (broader / more organized / easier creative thinking)
A single weight controls the mouth aperture:
(other example: single weight controls the swing of the stem of an apple)
