NeurIPS 2025 Explorer
Concepts
Authors
Glossary
johnsanterre.github.io
monosemantic features
3 papers
A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
Revising and Falsifying Sparse Autoencoder Feature Explanations