Kullback-Leibler (KL) Divergence

Asymmetric measure of difference between two probability distributions.

Neighborhood — ranked by edge-count

concept

KL Divergence
related_tosame_as
A measure of the difference between two probability distributions, used extensively in free energy formulations.

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.

KL Divergence Retention Evaluationmethod0.803
Measuring KL divergence between original and post-intervention outputs on Alpaca prompts to assess behavioral preservation
Representational Divergenceconcept0.787
Core phenomenon studied: when causal interventions shift internal representations away from the natural distribution
Pernicious Divergenceconcept0.721
Divergences that activate hidden pathways or cause dormant behavioral changes, undermining mechanistic claims
Paradigm Shift (Kuhn)framework0.720
Kuhn's concept that fundamental changes in basic assumptions are needed to move from one scientific/social paradigm to another.
Harmless Divergenceconcept0.713
Divergences that occur in the behavioral null-space and do not affect functional claims about the model
KL Controlframework0.712
Control framework minimizing expected complexity; shown to be a special case of expected free energy minimization
Kuhhandel (You’re Bluffing!)concept0.712
original tabletop card game designed by Rüdiger Koltze, licensed by Ravensburger
How can we produce a principled method for classifying harmful divergence for any mechanistic claim?question0.696
Identified gap: current work lacks a general method for harmful divergence classification