finding
active
finding:only-expressive-elementary-traits-creative-playful-in-both-models-passionate-in-q8b-are-steerable-care-oriented-traits-are-naturalOnly expressive elementary traits (creative/playful in both models, passionate in Q8B) are steerable; care-oriented traits are natural
Care-oriented elementary traits overlap with default helpful-assistant behavior, making them natural; expressive traits retain steering headroom
Source paper
extracted_from(2026) · Winston Zeng · Ali Emami · J H Choi
Neighborhood — ranked by edge-count
Papers (1)
paper
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.
- Q8B shows a broad steerable surface in the generic domain, like an uncommitted canvas
- Steering preferentially exposes exaggerated styles in G20B as well
- Key finding showing agentic behavior is encoded as a default operating mode in both models
- Both models independently converge on the same six clinician traits as natural defaults
- The costly error direction (labeling steerable as natural) almost never occurs
- Explains the distributional difference in generic domain: G20B has more mass at both extremes (natural and intractable)
- G20B has 7 steerable, 7 natural, 4 intractable generic traits, with more mass at extremes than Q8Bfinding0.762G20B's post-training commits more strongly toward and against particular dispositions, leaving fewer in the steerable middle
- Natural trait acts as anchor in mixed pairs, suppressing steerable partner at worst but never collapsing both