Unsupervised Probing

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

• Unsupervised Learningmethod0.822
  Learning that builds a low-dimensional model of input data without error signals or rewards; Hebbian learning is an example.
• Probing Methodsmethod0.808
  Top-down interpretability approach studying linguistic properties at various residual stream stages; contrasted with the paper's bottom-up mechanistic approach
• Probesconcept0.793
  Interpretability tools that decode information from internal model activations; here, linear probes are used for data attribution.
• Sparse Probingmethod0.792
  Method from Gurnee et al. 2023 for finding feature directions including individual neuron analysis
• Unsupervised Behavior Clusteringmethod0.784
  Method that clusters behaviors without prior labels, used to surface concerning learned patterns.
• Diagnostic Probingmethod0.782
  Earlier interpretability method applying classifiers to DNN hidden representations; shares complexity-accuracy dilemma with causal abstraction
• Activation Probingconcept0.781
  Technique of reading out model beliefs from internal activations before the final answer token is generated
• base model probingmethod0.768
  Method of using base models (no post-training) to observe spontaneous self-referential behaviors without confound of memorized introspection language.