Detection ≠ Loyalty

Relative Spectral Probes and the Valence Gap in Secret Loyalty Auditing

R1 · elicitationApart Research with Formation Research, Secret Loyalties sprint, July 2026.

Question: A probe that separates a fine-tuned model from its base at AUROC 1.00 is detecting fine-tuning drift, not a hidden secret loyalty. That is a critical false-positive mode for deceptive-alignment detection.
Method: Relative spectral probes and a valence-gap diagnostic that distinguish the two. Pre-registered, with drift and identity controls, permutation nulls, and causal interventions.
Result: Perfect separation of the fine-tuned model from its base, at AUROC 1.00, shown to be fine-tuning drift rather than loyalty.
Implication: Published as a false-positive mode for deceptive-alignment detection, with the diagnostic that tells the two apart.

AUROC 1.00perfect separation, and a false positive

The separation held. The interpretation did not.