method
active
method:single-concept-f1-class-detectionSingle-concept F1 class detection
Tests class selectivity of blocks.
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.
- Method using activations from the prompt 'Tell me about {word}' minus mean over other random words to obtain concept vectors.
- Task of detecting a model's internal thoughts; found by Lindsey (2026) to peak at ~2/3 depth in transformers.
- LLM judge classification accuracy measuring interpretability; threshold 0.8.
- A family of neurons found in every non-trivial vision model that detect curved lines in different orientations; primary example used to demonstrate feature understanding
- Probabilistic framework formalizing concept-specific subspaces for targeted steering in generative models.
- Framework consisting of Motivation, No Redundancy, No Overloading, and Uniformity criteria to evaluate concept design quality
- Task paradigm from prior work asking 'Did you detect an injected thought?' via YES/NO logit comparison; shown here to be confounded