concept
active
concept:inverse-scaling-on-disentanglement-metricsInverse Scaling on Disentanglement Metrics
Metrics worsen with dictionary size for most architectures.
Neighborhood — ranked by edge-count
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.
- Explains inverse scaling.
- Larger models behave worse on a metric; applied to faithfulness.
- Hypothesis cited in paper suggesting deceptive capabilities may scale with model size
- Matryoshka's hierarchical structure explains its positive scaling on disentanglement metricsclaim0.751Mechanistic interpretation.
- Cited hypothesis from Lin et al. 2022 suggesting larger models become more capable of deception
- Used in the color cooccurrence experiment to embed colors into 3D space preserving dissimilarity matrix distances
- CIMC's characterization of part of the solution to the Hard Problem: insight into the structural necessities of phenomenal representation
- Inverse SCR scaling persists across intervention sizes for non-hierarchical architecturesfinding0.733Rules out fraction-modified explanation.