Statistical Activation Analysis

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

• Activationsconcept0.768
  Internal representations of the model on which probes operate; the method uses activations to rank datapoints.
• Activation Correlationmethod0.766
  Pearson correlation of feature activations across 40M tokens used to measure feature similarity and universality across models
• Sequential SAE Activation Analysismethod0.757
  Token-level analysis of OTD and backtracking latent activations aligned at correction points across episodes
• Activation Additionmethod0.750
  Intervention method that adds a learned direction vector to residual stream activations to steer model behavior
• Activation Interval Samplingmethod0.749
  Dividing feature activation spectrum into 11 evenly-spaced intervals and sampling uniformly to evaluate monosemanticity across activation levels
• Activation Probingconcept0.740
  Technique of reading out model beliefs from internal activations before the final answer token is generated
• Statistical Physics (Statistical Mechanics)framework0.727
  The branch of physics dealing with large numbers of particles, statistical laws, and the tendency to disorder, as described by Boltzmann and Gibbs.
• Activation Similarityconcept0.725
  Model-independent feature comparison based on correlating activation vectors across a fixed diverse dataset