Reproducible tools and evidence for sparse autoencoder research
AI & ML interests
Building open foundations for understanding, evaluating, and intervening in neural networks.
Recent Activity
View all activity
Organization Card
Open science · Mechanistic interpretability
From learned features to testable explanations.
OpenSAE builds open research methods and infrastructure to study neural representations, intervene on model behavior, and evaluate how reliably our measurements explain it.
Research infrastructure
🧱 SAEBase Research Lab ↗Capture activations, intervene on SAE features, benchmark effects, and audit reliability.
▤ SAEBase Evaluation ↗Explore 18 model–SAE configurations and 48 verified model/layer sites with traceable reports.
Research methods and tutorials
models 0
None public yet