framework
active
framework:structured-sparsityStructured sparsity
Classical literature on sparsity priors over admissible supports.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Block-Sparse Featurizers (BSF)implementsFeaturizers whose atoms are blocks of directions rather than single directions.
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.
- Alternative sparsity notion proposed for predictive component.
- Property that features activate on only a small fraction of inputs; enables compressed sensing and is what allows superposition to work
- Training data with inherent geometric or relational structure, which induces geometric organization in model internals.
- The balance between how sparse and how faithful a decomposition is; VPD achieves a better tradeoff than transcoders.
- Structured sparsity question.
- SAEs assume sparsity (number of concepts) is time-invariant, drawn i.i.d. from a fixed distribution.claim0.751Corollary 4.1.1.
- Sparsity activation used with k=200.
- Higher sparsity forces composition of multiple concepts into fewer active latents, harming disentanglementhypothesis0.740Explains TPP/RAVEL preferring high L0.