OpenSAE

Team

non-profit

https://github.com/OpenSAE

Activity Feed

AI & ML interests

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

Recent Activity

Hunter26666666 authored a paper 38 minutes ago

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

Hunter26666666 authored a paper 38 minutes ago

Explaining AlphaGo: Interpreting Contextual Effects in Neural Networks

Hunter26666666 authored a paper 38 minutes ago

Rotation-Equivariant Neural Networks for Privacy Protection

View all activity

Papers

Behavioral Compositionality of Sparse Autoencoder Representations

View all Papers

Hunter26666666

authored 3 papers 38 minutes ago

updated a collection about 7 hours ago

OpenSAE Infrastructure & Evaluation

Collection

Reproducible tools and evidence for sparse autoencoder research • 2 items • Updated about 7 hours ago

Hunter26666666

updated 2 Spaces about 7 hours ago

OpenSAE

🧱

Open research for testable explanations of neural networks.

SAEBase Research Lab

🧱

Explore verified SAE checkpoints, interventions and results.

Hunter26666666

published a Space about 7 hours ago

SAEBase Research Lab

🧱

Explore verified SAE checkpoints, interventions and results.

Hunter26666666

updated a dataset about 7 hours ago

OpenSAE/SAEBase-Evaluation

Viewer • Updated about 7 hours ago • 240 • 1

Hunter26666666

published a dataset about 7 hours ago

OpenSAE/SAEBase-Evaluation

Viewer • Updated about 7 hours ago • 240 • 1

Hunter26666666

updated a collection about 8 hours ago

SAE Benchmarks, Datasets & Models

Collection

Benchmarks, datasets, and pretrained models for reproducible sparse autoencoder research, feature interventions, and mechanistic interpretability. • 10 items • Updated about 8 hours ago • 1

AI & ML interests

Recent Activity

Papers

Team members 1

OpenSAE's activity

OpenSAE

SAEBase Research Lab

SAEBase Research Lab