Litcius/Paper detail

InterPLM: discovering interpretable features in protein language models via sparse autoencoders

Elana P. Simon, James Zou

2025Nature Methods33 citationsDOI

Topics & Concepts

Computer scienceArtificial intelligenceFeature (linguistics)Interpretation (philosophy)Language modelMachine learningSequence (biology)Natural language processingLanguage understandingPattern recognition (psychology)InterpretabilitySuperposition principleComputational linguisticsSequence labelingAnnotationMechanism (biology)Topic ModelingMachine Learning in BioinformaticsNatural Language Processing Techniques
InterPLM: discovering interpretable features in protein language models via sparse autoencoders | Litcius