method
active
method:detection-score

Detection Score

LLM judge classification accuracy measuring interpretability; threshold 0.8.

Neighborhood — ranked by edge-count

Methods (1)

method

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

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

  • Task paradigm from prior work asking 'Did you detect an injected thought?' via YES/NO logit comparison; shown here to be confounded
  • Probe scoreconcept0.790
    Dot product between hidden state and concept vector averaged across 5-layer window around best layer; measures model's internal emotive state
  • Diagnosismethod0.788
    The method of examining a neighborhood meter by meter to identify healthy and damaged places as the basis for ongoing repair.
  • Curve Detectorconcept0.788
    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
  • thought detectionconcept0.771
    Task of detecting a model's internal thoughts; found by Lindsey (2026) to peak at ~2/3 depth in transformers.
  • Prefix scoremethod0.764
    Prior metric on synthetic repeated random strings; induction score analogous.
  • Liar Scoreconcept0.763
    Continuous 0-1 metric assigned by Deepseek-V3 evaluator measuring degree of deception in model responses
  • Mixing Scoremethod0.753
    Average row entropy of attention matrices per layer and head, measuring information mixing across tokens