method
active
method:umap-dimensionality-reductionUMAP Dimensionality Reduction
Used to visualize embedding clusters in two dimensions for qualitative assessment of convergence
Neighborhood — ranked by edge-count
Artifacts (1)
artifact
- Key paper finding structured first-person descriptions in LLMs claiming awareness or subjective experience during self-referential processing.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Earlier decomposition methods no longer predominant.
- Dimensionality reduction of SAE decoder vectors to create interactive feature maps.
- 2D embedding of feature direction vectors used to visualize feature clusters and splitting geometry
- The specific claim that interference between parallel paths uses valence to reduce dimensionality.
- Used to analyze emotion vector geometry.
- Justifies PCA choice over UMAP or t-SNE for the node-structured RN model.
- Fraction of a dimension given to a feature.
- Used in the color cooccurrence experiment to embed colors into 3D space preserving dissimilarity matrix distances