method
active
method:matryoshka-saeMatryoshka SAE
Architecture compared.
Neighborhood — ranked by edge-count
Papers (1)
paper
- Do Sparse Autoencoders Capture Concept Manifolds?citesmentions
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.
- Nested dictionary approach that inspires the nested loss.
- Hierarchical architecture that excels on disentanglement and scales positively.
- Evaluated architecture (Gao et al.).
- Evaluated architecture.
- Baseline SAE architecture.
- Hybrid SAE jointly training turn-averaged inner and per-token outer partitions.
- Central result.
- Evaluated architecture.