finding
active
finding:on-cifar-10-larger-models-exhibit-greater-alignment-with-each-other-compared-to-smaller-onesOn CIFAR-10, larger models exhibit greater alignment with each other compared to smaller ones
Kornblith et al. / Krizhevsky finding replicated in paper discussion
Source paper
extracted_from(2024) · Minyoung Huh · Brian Cheung · Tongzhou Wang · Phillip Isola
Neighborhood — ranked by edge-count
Hypotheses (1)
hypothesis
- Selective pressure toward convergence via model capacity
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.
- Scale-dependent structural finding from PCA visualizations in §4
- Concurrent work result showing emergent misalignment occurs in small models
- Latent #10 activation increase correctly classifies all correct vs incorrect fine-tuned models in Figure 9
- Scale-dependent alignment result demonstrating how more abstract truth representations emerge with scale
- Larger rotations between shared features skew decoder cosine similarities downward in the toy modelfinding0.754Supports 'same features, used differently' interpretation.
- Experiment 4 result showing DIM captures only one facet of the multi-dimensional truth subspace
- Extrapolation from scale-emergence finding to future risk