Comment by Berkan Ottlik (berkan-ottlik) on 200 COP in MI: Image Model Interpretability · 2023-01-09T17:51:50.328Z · LW · GW

Different image model architectures

I've done a little bit of work on ViT interpretability. It's kind of messy right now but maybe a starting point for someone else to jump off of + I might add to it in the future: https://berkan.xyz/projects/ (see vision transformer interpretability).