arxiv
PublishedOctober 2, 2026 at 4:00 AM
—neutral
Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout
Publisher summary· verbatim
arXiv:2609.39263v1 Announce Type: new Abstract: A concept subspace's effect on model behavior does not establish how it relates to the output readout. We introduce a two-sided geometric diagnostic that measures an extracted subspace's overlap with the dominant right-singular directions of the unembe
Stay posted· Newsletter
Three short emails a week: Monday, Wednesday and Friday. Top movers, new models and the most-referenced story.
Your email is used only to send The Bubble Brief. Privacy policy
Discussion
No replies yet. Be first.
The Bubble Brief
MON · WED · FRIRead AI insights every Monday, Wednesday and Friday — top movers, new models, story of the week.
Your email is used only to send The Bubble Brief. Privacy policy
Originally published on arxiv ↗