Recognition Density

Approximate posterior probability distribution embodied in organism's internal states; organism's best guess about causes of sensations

Neighborhood — ranked by edge-count

claim

Generative models are not structural representations; recognition densities are
about
Direct refutation of structural representationalist interpretation; recognition density encodes information, not generative model

concept

free energy
associated_with
Thermodynamic potential ΔF = ΔE − TΔS; domain walls form if ΔF < 0
Embodied Inference
associated_with
Process by which organism's material states and internal dynamics realize variational inference through action

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.

Feature Densityconcept0.804
Fraction of training tokens on which a given feature has nonzero activation; used as proxy metric for autoencoder quality
Burden of Recognitionconcept0.768
Waldron's civility requirement to recognize opposing views as genuine opposing views rather than products of irrationality.
Feature Density Histogrammethod0.741
Log-scale histogram of feature firing rates used as proxy for autoencoder quality during hyperparameter tuning
Density of Centersconcept0.730
The degree to which centers are packed and overlapped, contributing to the life of the whole.
Representational Divergenceconcept0.723
Core phenomenon studied: when causal interventions shift internal representations away from the natural distribution
Natural Distribution of Representationsconcept0.721
The distribution of latent representations produced by the model under unperturbed inputs
Perceptionconcept0.719
Equated with inference of past, present and future hidden states via minimization of variational free energy.
Data Attributionconcept0.719
The task of attributing model behaviors to specific training datapoints.