In Gemma 2 27B, just 4 PCA components explain 70% of the variance among 275 character archetype roles (vs Qwen 3 32B needing 8, Llama 3.3 70B needing 19)

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

• 4-19 principal components explain 70% of variance in role persona space across the three models (Gemma 4, Qwen 8, Llama 19)finding0.835
  Demonstrates that persona space is low-dimensional
• Pairwise correlation of role loadings on PC2 is 0.89 between Qwen and Llama; Gemma differs (similarity <0.61) from others on PC2finding0.819
  Characterizes model similarities and differences in secondary persona dimensions
• Persona space is remarkably low-dimensional: just 4 components explain 70% of variance among 275 roles in Gemma 2 27Bclaim0.817
  Empirical support for Hypothesis 2, suggesting tractable structure in the space of LLM personas
• Base and instruct Gemma 2 27B role PCs have cosine similarities of 0.93, 0.87, 0.83 for the top 3 PCs respectively; role vector cosine similarities >0.99 for every role pairfinding0.806
  Shows persona space axes are inherited from pre-training, not solely created by post-training
• Pairwise correlation of role loadings on PC1 exceeds 0.92 across all model pairs, indicating remarkably high similarity of the Assistant Axis across Gemma, Qwen, and Llamafinding0.784
  Shows the leading component of persona space is model-universal
• Gemma 2 27B is unlikely to take on human personas when steered away from Assistant, preferring nonhuman or theatrical portrayalsfinding0.779
  Model-specific difference in persona susceptibility
• Heads 24, 30, and 32 in layer 14 of Llama-3.1-8B consistently emerged as top contributors for every evaluated persona traitfinding0.777
  Identifies specific Style Modulation Heads in Llama-3.1-8B
• Gemma's Assistant appears emotionally regulated and systematic; Qwen appears pedagogical and thoughtful; Llama appears socially intelligent and warmclaim0.770
  Model-specific characterizations of what the Assistant persona looks like across different models