Query Lens extends Logit Lens to interpret sparse features via key-value analysis and indirect effects, yielding coherent token signatures where Logit Lens fails, and proposes the Subspace Channel Hypothesis.
Enhancing Automated Interpretability with Output-Centric Feature Descriptions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects
Query Lens extends Logit Lens to interpret sparse features via key-value analysis and indirect effects, yielding coherent token signatures where Logit Lens fails, and proposes the Subspace Channel Hypothesis.