Litcius/Paper detail

Towards Automated Circuit Discovery for Mechanistic Interpretability

Arthur Conmy, Adrià Garriga-Alonso, Stefan Heimersheim, Aengus Lynch, Augustine N. Mavor-Parker

20237 citationsDOI

Topics & Concepts

InterpretabilityComputer scienceArtificial intelligenceFeature (linguistics)Machine learningData miningKey (lock)Pattern recognition (psychology)Class (philosophy)Identification (biology)Machine Learning and AlgorithmsExplainable Artificial Intelligence (XAI)VLSI and FPGA Design Techniques
Towards Automated Circuit Discovery for Mechanistic Interpretability | Litcius