LLMs Faithfully and Iteratively Compute Answers During CoT: A Systematic Analysis With Multi-step Arithmetics
Uses linear probes and activation patching on multi-step arithmetic to test when answers are computed. Finds that the tested models form intermediate answers during written CoT and that those generated steps causally affect the final answer.
First public: · Added to this list: November 9, 2026
Explore this approach
Analysis and evaluation — Measure how well these methods work, investigate what their internal computation does, and identify where they fail. This includes experiments, studies of model internals, and mathematical results.