concept
active
concept:features-mechanistic-interpretabilityFeatures (mechanistic interpretability)
Functionally significant directions in the residual stream that organize activations in mechanistic interpretability
Neighborhood — ranked by edge-count
Papers (1)
paper
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.
- Explaining activations without understanding mechanism.
- Goodfire's stance drawn from experience with Evo and R1.
- An inference path across features that constitutes a computational subgraph in a trained model
- The capability to explain model predictions; a central theme of the paper, with disruption profiles as vehicle.
- New affordance of features beyond monitoring/steering.
- Model organism choice.
- Socio-philosophical open question.