awesome-interpretability
github.com/wassname/awesome-interpretability ↗Awesome tools for interpreting, manipulating the internals of of deep neural networks.
19
GitHub Stars
152
Curated Resources
18
Categories
4 days ago
Last Refreshed
Inspect and interveneNatural-language interpretationLenses, probes and conceptsCausal hypotheses and circuit discoverySteering and trainable interventionsModel organismsEvaluate interpretations and interventionsInterpretable models by designExplainability and counterfactualsSparse feature toolingBrowsers and visualizationsAgent auditing and evaluation infrastructureAdaptersMine (wassname)Structured output (adjacent tooling)PerspectivesReading and tutorialsSee more
Use this list with your AI agent
Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:
"Show me model organisms resources from awesome-interpretability"
Installation instructions →What's inside
Model organisms
Mine (wassname)
Natural-language interpretation
Adapters
Perspectives
- AGI Ruin #27The Most Forbidden Technique
Eliezer Yudkowsky; Zvi Mowshowitz
- A Pragmatic Vision for Interpretability
Neel Nanda and the GDM interpretability team
- Chain of Thought Monitorability: A New and Fragile Opportunity
Tomek Korbak, Mikita Balesni and 39 co-authors
- SimulatorsThe Persona Selection Model
janus; Sam Marks, Jack Lindsey, Chris Olah
- The case for ensuring that powerful AIs are controlled
Ryan Greenblatt, Buck Shlegeris
- The Urgency of Interpretability
Dario Amodei
Reading and tutorials
Explainability and counterfactuals
Causal hypotheses and circuit discovery
Showing a sample of 152 resources. View the full list on GitHub →