Minimum Description Length Probing

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

• Minimum Description Lengthframework0.867
  Related to variational free energy; compressibility corresponds to complexity reduction in structure learning
• Diagnostic Probingmethod0.760
  Earlier interpretability method applying classifiers to DNN hidden representations; shares complexity-accuracy dilemma with causal abstraction
• Sparse Probingmethod0.750
  Method from Gurnee et al. 2023 for finding feature directions including individual neuron analysis
• Probing Methodsmethod0.737
  Top-down interpretability approach studying linguistic properties at various residual stream stages; contrasted with the paper's bottom-up mechanistic approach
• Linear Probingmethod0.735
  Used to evaluate representation quality across VTAB tasks
• Unsupervised Probingmethod0.725
  Probing approach avoiding supervision to sidestep complexity-accuracy tradeoff
• base model probingmethod0.724
  Method of using base models (no post-training) to observe spontaneous self-referential behaviors without confound of memorized introspection language.
• Probesconcept0.711
  Interpretability tools that decode information from internal model activations; here, linear probes are used for data attribution.