claim
active
claim:tiling-of-curve-detectors-is-the-predictable-optimum-of-a-mismatched-flat-sparse-priorTiling of curve detectors is the predictable optimum of a mismatched flat sparse prior.
Retrodiction of known phenomenon.
Source paper
extracted_from(2026) · Thomas Fel · Matthew Kowal · Mozes Jacobs · Dron Hazra +21
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Multiplicity of SAE curve features is an artifact of the featurizer, not a property of the network.supportsReinterpretation of curve detectors.
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.
- Appendix B.
- Speculation that QK circuit 'concordance heads' underlie the ability to distinguish intended from unintended outputs.
- Curve detectors found across AlexNet, InceptionV1, VGG19, ResNetV2-50 and models trained on Places365finding0.757Anecdotal evidence for the universality of low-level visual features across different architectures and datasets
- Smoothed behavior.
- Thought detection peaks at ~2/3 layer depth; intention checking peaks at ~1/2 layer depth.finding0.755Lindsey (2026) differential layer performance explained by Janus's path combinatorics — different tasks use different path distributions.
- Demonstrates that meaningful algorithms can be read directly off floating-point weights in a neural network
- Crosscoder model diffing is best used as a high-recall pre-screening tool followed by verification.claim0.752Recommended workflow.