concept
active
concept:model-organisms-for-interpretabilityModel organisms for interpretability
Proposal for standard studied networks akin to Drosophila.
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Proposal.
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.
- A model deliberately trained to exhibit alignment-relevant properties so researchers can study them with ground truth.
- The capability to explain model predictions; a central theme of the paper, with disruption profiles as vehicle.
- Socio-philosophical open question.
- Frames model interpretability like biology, features as cells, tools as microscopes.
- Probability of data under the model, penalizing complexity and rewarding accuracy.
- Motivation for the release.
- Proposed paradigm for evaluating interpretability work through empirical falsifiability rather than benchmarks or user studies
- Opening question.