Anthropic Global Workspace Paper Uses Jacobian Lens for LLM Interpretations
Reactions from ranked influencers
4 posts
enjoying the new global workspace anthropic paper - especially using the Jacobian lens to provide readout interpretations of prints shown at last month CVPR
previously my main intuition pump nearest neighbors in vision models (shown here). my goal is always to make drawings that generalize to new models and techniques so happy that understanding also works here. amusingly if you go upstream responses simplify to just "books books…
here's the same effect with a different print: on the 48 layer gemma3-12b, at layer 47 you get bird, woodpecker, ornothology but upstream at layer 37 its again just "bird birds birds"
just some early results - slightly cherry-picked, but things seem to improve with scale which is a good sign. kudos to @wesg52 @sofroniewn @Jack_W_Lindsey and team for the refrence implmentation which made it easy to explore! https://x.com/AnthropicAI/status/2074185348142280912