finding
active
finding:turn-averaged-features-have-highest-embedding-similarity-0-663-at-k-3-per-token-max-lowest-at-0-617

Turn-averaged features have highest embedding similarity 0.663 at k=3; per-token max lowest at 0.617

Consistent across k=1,2,3,5.

Source paper

extracted_from
Turn-Averaged SAEs for Feature Discovery and Long-Context Attribution
(2026) · Kevin Der · Harish Kamath · Thompson, Ben

Neighborhood — ranked by edge-count