Researcher argues DoM steering is no more transparent than gradient-based vector selection
johnhewtt · x · 2026-10-27
In a discussion with @arunasank, the author argues that DoM (Direction of Motivation) is no more transparent than gradient-based selection of steering vectors — both are ultimately interpreted nonlinearly by a deep network.
His key point: localization (possibly wrong) and the activation-editing method DoM (also possibly wrong) are separate components, and the finding that DoM works better at a layer chosen by localization doesn't validate either method.
Related event: Interpretability debate: steer by difference of means or by gradients?(6 posts)→
More from Research
- Kevin Bryan: maximizing raw accuracy is rarely the right way to train AI assistants — Afinetheorem · 2026-10-27
- Szegedy: a flaw in common training losses went unnoticed for over a decade — ChrSzegedy · 2026-10-27
- Claude research model claims to refute Exponential Time Hypothesis with 14M-token autonomous run — scaling01 · 2026-10-27
- "ETH is false": CS community buzzing over claim that upends SAT exponential-time assumption — akbirthko · 2026-10-27
- RL environments essay: knowing how to operate software isn't knowing how to achieve goals — ShenRaphael · 2026-10-27
- Δτ-Bench: Agents Face a Moving Tool Frontier, Best Model Scores Just 42.3% — UniversityCollegeLondon · 2026-10-27