claim
active
claim:sparsity-is-not-a-good-proxy-for-interpretabilitySparsity is not a good proxy for interpretability
SDL assumption may break down.
Source paper
extracted_from(2025) · Lee Sharkey · Bilal I. Chughtai · Joshua Batson · Jack W. Lindsey +25
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- Feature splitting and absorptionsupportsProblems suggesting sparsity breaks down as interpretability proxy.
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.
- Better bases may give sparsity.
- Benchmark discipline.
- A better basis (e.g., factoring tokens into languages or parts of speech) may yield a sparse interpretable modelhypothesis0.783Same model can be dense in one basis and sparse in another.
- Caveat on metric.
- Ian Goodfellow quote used to illustrate the pre-paradigmatic state of interpretability research
- Motivates multi-metric evaluation.
- Interpretability hypotheses are often conflated with conclusions and require extensive validationclaim0.762Validation is under-done.
- Proxy argument.