PaperZoom In
Zoom In
A landmark 2020 mechanistic interpretability paper by Chris Olah and colleagues that used visualization analysis to reveal the phenomenon of feature superposition in neural networks, providing key evidence for the superposition hypothesis.
Timeline (last 90 days)
Oct 16
In 2020, Cammarata and Chris Olah provided evidence in the paper "Zoom In" showing that the network hides car feature signals inside a group of dog neurons, reinforcing the superposition hypothesis.
Unverified50%