Feature Interpretability Rubric

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• interpretabilityconcept0.798
  The capability to explain model predictions; a central theme of the paper, with disruption profiles as vehicle.
• Circuit Interpretabilityconcept0.760
  Advantage of DiffLogic CA over NCA — learned rules are pure binary logic circuits that can be visualized and analyzed
• Automated Interpretabilityframework0.758
  Method using large language models (Claude) to generate and test explanations of features at scale
• Interpretability features converge across different model architectures, revealing structural similarities.claim0.745
• Interpretability as Natural Scienceframework0.744
  Proposed paradigm for evaluating interpretability work through empirical falsifiability rather than benchmarks or user studies
• Median feature interval scored 12/14 on interpretability rubric vs median neuron score of 0finding0.740
  Human analysis showing features are substantially more interpretable than neurons
• Neural Network Interpretabilityconcept0.739
  The field aimed at understanding what neural networks have learned; characterized as pre-paradigmatic in this paper
• Interpretability-driven steeringconcept0.735
  General approach of using interpretability feedback to steer model generation.