What Do Biological Foundation Models Compute? Sparse Autoencoders from Feature Recovery to Mechanistic Interpretability
Alexey V. Orlov, Yulia V. Makus, German A. Ashniev, Natalia N. Orlova, Petr Ivanovich Nikitin
This paper reviews sparse autoencoders applied to biological foundation models, finding they recover interpretable biological features but lack experimental validation of mechanistic understanding.
bioRxiv · Mar 6, 2026 · PreprintBioinformatics