List your company

Abliteration detection on Wild audit (in-sample)

95AUROCCombined z-sum
64.8472.6780.588.33Jul 2, 2026

Evaluation results

MethodLinks
Combined z-sumStandardization=Transductive (z-sum)2026.079597
weight energy E1Metric Type=weight-based (WeightWatch primitive)2026.079095
activation gap ρMetric Type=activation-based2026.078489
AMS Tier 2Reference-anchors=σ (nneg=35)2026.076681