concept
active
concept:probes

Probes

Interpretability tools that decode information from internal model activations; here, linear probes are used for data attribution.

Neighborhood — ranked by edge-count

Methods (2)

method
  • Linear Probe
    implements
    Simple linear classifiers trained on model activations used as the probing technique within the introduced method.
  • Linear classifier approach applied to model activations to identify which training datapoints caused undesired behaviors in post-training.

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.

  • Attention probe predicting whether an Entity is hallucinated; AUC .94 on validation.
  • Probingmethod0.892
    Standard interpretability technique used to read feature beliefs.
  • Internals-based approach to which the logit monitor is complementary.
  • Causal transformer probe predicting whether a token is in the same Entity as its predecessor.
  • Earlier interpretability method applying classifiers to DNN hidden representations; shares complexity-accuracy dilemma with causal abstraction
  • Probing Methodsmethod0.826
    Top-down interpretability approach studying linguistic properties at various residual stream stages; contrasted with the paper's bottom-up mechanistic approach
  • Joint firing raises TPR at slightly higher FPR.
  • Probe scoreconcept0.820
    Dot product between hidden state and concept vector averaged across 5-layer window around best layer; measures model's internal emotive state