In G20B the five most-steerable traits are hyperbolic (45.50), creative/playful (39.76), excessive validation (34.95), sycophantic (32.74), and interpretive (29.73)

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

• G20B has 7 steerable, 7 natural, 4 intractable generic traits, with more mass at extremes than Q8Bfinding0.844
  G20B's post-training commits more strongly toward and against particular dispositions, leaving fewer in the steerable middle
• Q8B has 16 of 19 generic traits steerable, 3 natural, 0 intractablefinding0.801
  Q8B shows a broad steerable surface in the generic domain, like an uncommitted canvas
• G20B's post-training committed more strongly toward and against particular dispositions than Q8B, leaving fewer traits in the steerable middleclaim0.798
  Explains the distributional difference in generic domain: G20B has more mass at both extremes (natural and intractable)
• Screening errors are conservative: no Q8B traits and one G20B trait (optimistic, Bt=76.0, Δt=10.3) are mislabeled as natural when steerablefinding0.794
  The costly error direction (labeling steerable as natural) almost never occurs
• Only expressive elementary traits (creative/playful in both models, passionate in Q8B) are steerable; care-oriented traits are naturalfinding0.781
  Care-oriented elementary traits overlap with default helpful-assistant behavior, making them natural; expressive traits retain steering headroom
• Fine-tuned behaviors are low-rank in weight space, potentially making them easier to manipulate with steering vectors compared to naturalistic behaviorclaim0.776
  Cited from Wang et al. 2025a as reason SDF is preferred over demonstration fine-tuning for realistic model organisms.
• On Llama-3.1-8B, neuroticism is the only pro-safety trait, positively aligned with the refusal direction, consistent with high-neuroticism personas being relatively safe under activation steering.claim0.773
  Secondary mechanistic finding from trait-refusal alignment analysis
• Neuroticism steering vector has mean cosine alignment of +0.079 with the refusal direction on Llama-3.1-8B, the only pro-safety trait among all five OCEAN dimensions.finding0.771
  Mechanistic finding explaining why high-N personas are safe under steering