Activation-Space Order-Swap Geometry: A Site-Asymmetry Audit

The paper introduces a site-asymmetry audit for activation-space interventions, decomposing order-dependent activation statistics into a canonical additive response from single interventions and an antisymmetrized second difference that removes first-order and self-curvature effects. Across six language-model families, the single-intervention baseline accounts for most of the bracket norm, while the corrected residual often clears a generic-interaction null. The method also transfers to non-language models, demonstrating its portability.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.