List your company

SAE Feature Interpretability on Gemma SAE Features Value Output 3-1B

5.3Interpretability ScoreQuery Lens
4.55124.74564.945.1344May 30, 2026

Evaluation results

MethodLinks
Query Lensvariant=QL_VALUE2026.055.3
Logit Lensvariant=LL_VALUE2026.054.58