PCA Latent Space Trajectory

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

• PCA on Persona Spacemethod0.766
  Standardized PCA run on role vectors to find main axes of persona variation
• Principal component analysis of persona spacemethod0.758
  Method used by Lu et al. to find orthogonal directions of maximum variance among 275 character archetypes in activation space
• PCA Visualizationmethod0.756
  Used to visually inspect separation of truth-related directions in model activation space across layers
• Principal components analysis (PCA)method0.742
  Statistical method used to analyze neural activity data.
• Faster-converging trajectories take direct PCA-space paths to the true solution; slower trajectories take indirect routes visiting saddle pointsfinding0.739
  PCA-space visualization linking route geometry to convergence speed.
• Latent-Space Representationsconcept0.732
  Substrate on which causal emergence was computed across agent lifetimes; aligned with learning success.
• The first principal component of persona space (the 'Assistant Axis') captures the assistant vs alternative persona distinction consistently across three models with role loading correlations exceeding 0.92finding0.714
  Finding establishing cross-model consistency of the assistant axis as the dominant structure in persona space
• Principal Component Analysis Visualizationmethod0.708
  Used to visualize LLM true/false representations, revealing clear linear structure separating true from false statements