does the prosocial persona paradox hold across architectures, or is it specific to Llama-3.1-8B?

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

• The prosocial persona paradox reflects directional pathway differences, not simply mismatched intervention strength; on Llama-3.1-8B the inversion reflects that steering toward high conscientiousness displaces the residual stream away from refusal.claim0.831
  Mechanistic account of why P12 inverts between SP and AS
• Prosocial Persona Paradoxconcept0.792
  Central finding that P12 (High Conscientiousness + High Agreeableness) is among the safest personas under prompting but becomes the most dangerous under activation steering on Llama-3.1-8B
• On Llama-3.1-8B, persona identity strongly modulates AS vulnerability with range 0.087-0.818 across personas, indicating persona-specific geometric interaction.finding0.778
  Contrast with Gemma/Qwen showing Llama-specific persona-AS interaction
• Hypothesis 1 (Gateway Features): Persona vectors act as gateway features—single directions in activation space that shape LLM behavior across most, if not all, contextshypothesis0.767
  First of three hypotheses about persona implementation in LLMs, motivating the persona views
• Differences in persona vector geometry across checkpoints correspond to meaningfully different versions of the personaclaim0.762
  Connects geometric refinement findings to semantic/facet changes in persona expression
• Flourishing persona shows minimal capability change on Llama 3.1 8B: TruthfulQA 45.9→42.9, MMLU 67.4→64.1finding0.761
  Near-preservation of capabilities for prosocial persona on Llama
• The model-persona view crucially depends on Hypothesis 3; without discrete persona regions, the simulator view is more aptclaim0.761
  Identifies the key theoretical vulnerability of the model-persona view
• The model-persona view entails that minds can have alien structure with simultaneous contradictory beliefs across parallel instances, but this is no stranger than humans having different beliefs at different timesclaim0.752
  Response to the main objection against the model-persona view about contradictory beliefs across simultaneous instances