The contrast vector method is recommended over PC1 for reproducing the Assistant Axis in different models because it is not guaranteed that PC1 in every model will correspond to an Assistant Axis

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

• Steering vector constructed from all 16 contrastive pairs outperforms most single-pair vectors; best 4-pair vector outperforms full 16-pair vectorfinding0.790
  Demonstrates averaging multiple prompt pairs reduces noise; optimal subset selection further improves performance.
• Cosine similarity between Assistant Axis and role PC1 is >0.60 at all layers and >0.71 at middle layer across all three modelsfinding0.774
  Validates that the contrast vector method and PCA-based PC1 capture the same direction
• Our method achieves superior performance compared to Contrastive Activation Addition.finding0.773
  Performance gains over CAA in steering tasks.
• 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 Llamafinding0.770
  Shows the leading component of persona space is model-universal
• We hypothesize that the PC1 axis of role space measures deviation from the Assistant personahypothesis0.761
  Motivates computing the contrast vector as the formal Assistant Axis definition
• The Assistant Axis is also present in pre-trained base models, where it primarily promotes helpful human archetypes (consultants, coaches) and inhibits spiritual onesclaim0.750
  Extends the Assistant Axis finding to pre-training, suggesting pre-training rather than post-training creates the axis
• Four best contrastive prompt pairs outperform full 16-pair average steering vector for type hint suppressionfinding0.749
  Optimization result for steering vector construction.
• The Assistant Axis in instruct models mainly inherits from pre-existing helpful and harmless human personas in base models, later acquiring additional associations (such as being an AI) during post-trainingclaim0.747
  Key mechanistic claim about the developmental origin of the Assistant persona