OpenSAE

AI & ML interests

Building open foundations for understanding, evaluating, and intervening in neural networks.

Recent Activity

Hunter26666666  authored a paper about 14 hours ago

Quantification and Analysis of Layer-wise and Pixel-wise Information Discarding

Hunter26666666  authored a paper about 14 hours ago

Explaining AlphaGo: Interpreting Contextual Effects in Neural Networks

Hunter26666666  authored a paper about 14 hours ago

Rotation-Equivariant Neural Networks for Privacy Protection

View all activity

Papers

Behavioral Compositionality of Sparse Autoencoder Representations

View all Papers

Team members 1