Dormant Subspace

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Subspace DASmethod0.757
  Extension of DAS that learns a second rotation matrix on top of a fixed first one to decompose representations into sub-representations.
• Dormant Behavioral Changesconcept0.746
  Perturbations behaviorally null in one context but altering behavior in another due to latent divergence
• Behaviorally Binary Subspaceconcept0.741
  A vector subspace that causally impacts outputs only through the sign of its values, enabling harmless magnitude divergence
• Truth Subspaceconcept0.739
  The multi-dimensional activation subspace whose directions causally mediate truthful behavior in LLMs
• Subspace Interventionconcept0.727
  Intervention targeting specific dimensional subsets of activation vectors rather than full representations
• Balanced Subspacesconcept0.725
  Subspaces whose contributions to a layer's output are canceled by opposing weight values, making them non-causally active under natural inputs
• Emotion Subspaceconcept0.725
  The subspace of activation space spanned by the 171 orthogonalized emotion probe vectors, used to measure SAE feature emotional alignment
• Subspace Decomposition of Representationsconcept0.725
  Investigation of whether a distributed representation can be further decomposed into sub-representations encoding component identities.