Mechanistic Interpretability Via Learning Differential Equations: AI Safety Camp Project Intermediate Report.
by Valentin2026, ayoakin, Eduard Kovalets, tz3r0n4r, alytose, Utkarsh Priyadarshi, Varun Piram, and Axel Ahlqvist
TLDR; We report our intermediate results from the AI Safety Camp project “Mechanistic Interpretability Via Learning Differential Equations”. Our goal was to explore transformers that deal with time-series numerical data (either infer the governing differential equation or predict the next number). As the task is well formalized, this seems to...
May 8, 20258