method
active
method:subspace-dasSubspace DAS
Extension of DAS that learns a second rotation matrix on top of a fixed first one to decompose representations into sub-representations.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- Distributed Alignment SearchextendsThe core method introduced in this paper: finds alignments between high-level causal variables and distributed neural representations via gradient descent.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The subspace of activation space spanned by the 171 orthogonalized emotion probe vectors, used to measure SAE feature emotional alignment
- Intervention targeting specific dimensional subsets of activation vectors rather than full representations
- The multi-dimensional activation subspace whose directions causally mediate truthful behavior in LLMs
- Burger et al. (2024) framework proposing that truth is linearly decoded along a 2D subspace capturing both polarity-dependent and polarity-invariant directions.
- Subspaces whose contributions to a layer's output are canceled by opposing weight values, making them non-causally active under natural inputs
- A vector subspace that causally impacts outputs only through the sign of its values, enabling harmless magnitude divergence
- Investigation of whether a distributed representation can be further decomposed into sub-representations encoding component identities.
- Subspace dimensions that do not vary between inputs; included in the extraneous subspace zextra.