Skip to content
LessWrong AI · Communities

Can we teach a model to encode a semantic feature on a chosen manifold in just three channels?

This is my submission to BlueDot's Technical AI Safety Puzzle #1, for which I received an Honorable Mention. Congratulations to Gustavo Korzune Gurgel, Patryk Perduta (his amazing write-up), Sam Spilllard, Karine Levonyan, and Michael Zlatin for their recognition in the puzzle.My article below focuses on my answer to T