List your company

Circuit Recovery on MCQA

1.13CPR (Faithfulness Score)NAP-IG-Inputs
0.21480.45240.690.9276Apr 27, 2026

Evaluation results

MethodLinks
NAP-IG-InputsMethod Category=NAP-IG-Inputs, Transfer Setting=Gold (Upper Bound)2026.041.13
NAP-IG-ActivationsMethod Category=NAP-IG-Activations, Transfer Setting=Gold (Upper Bound)2026.040.9
NAPMethod Category=NAP, Transfer Setting=Gold (Upper Bound)2026.040.82
NAP-IG-InputsTransfer Setting=DFA (Far/Zero-Shot), Model Pair=Llama-3 (1B -> 3B)2026.040.48
NAP-IG-InputsTransfer Setting=Gold (Upper Bound), Model Pair=Llama-3 (1B -> 3B)2026.040.47
NAP-IG-InputsTransfer Setting=DFA (In-Distribution), Model Pair=Llama-3 (1B -> 3B)2026.040.41
NAP-IG-InputsMethod Category=NAP-IG-Inputs, Transfer Setting=DFA (Far/Zero-Shot)2026.040.33
NAP-IG-InputsMethod Category=NAP-IG-Inputs, Transfer Setting=DFA (In-Distribution)2026.040.32
NAP-IG-ActivationsMethod Category=NAP-IG-Activations, Transfer Setting=DFA (Far/Zero-Shot)2026.040.3
BaselineMethod Category=Baseline, Transfer Setting=Random Alignment2026.040.28
NAPMethod Category=NAP, Transfer Setting=DFA (Far/Zero-Shot)2026.040.28
NAPMethod Category=NAP, Transfer Setting=DFA (In-Distribution)2026.040.26
NAP-IG-ActivationsMethod Category=NAP-IG-Activations, Transfer Setting=DFA (In-Distribution)2026.040.25
BaselineTransfer Setting=Random Alignment, Model Pair=Llama-3 (1B -> 3B)2026.040.25