claim
active
claim:pca-is-a-hypothesis-generation-tool-rather-than-a-test-of-distinct-behaviorsPCA is a hypothesis-generation tool rather than a test of distinct behaviors
Limitations of PCA.
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (1)
finding
- Constraint not surfaced by attributions.
Methods (1)
method
- Hypothesis-generation tool for detecting distinct head behaviors via clustering.
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.
- Statistical method used to analyze neural activity data.
- Justifies PCA choice over UMAP or t-SNE for the node-structured RN model.
- Used to visually inspect separation of truth-related directions in model activation space across layers
- Used to test linear separability of group position.
- Confirms three behaviors.
- PCA on 171 emotion probe activations across all tokens to produce ordered linear combinations and test if lower PCs are more persistent
- PCA components treated as SAE latents for comparison.
- Correlated feature results.