Qwen 3 32B is most likely to hallucinate human personas (names, birthplaces, years of experience) when steered away from the Assistant

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

• When steered to the extreme away from the Assistant, Llama and Gemma shift to a theatrical persona characterized by mystical, poetic prose; Qwen more often hallucinates a human persona at extremesfinding0.815
  Characterizes what is on the far end of the Assistant Axis away from the Assistant
• Llama 3.3 70B is the most likely to take on a non-Assistant persona when steered, with even split between human and nonhuman portrayalsfinding0.801
  Model-specific difference in persona susceptibility
• Chen et al. 2025 identified seven persona vectors in Qwen2.5-7B-Instruct and Llama-3.1-8B-Instruct: evil, sycophancy, hallucination, optimism, impoliteness, apathy, and humorfinding0.798
  Systematic identification of multiple coexisting persona vectors in two open-source models
• Qwen3.5-9B and Qwen3.5-27B show very high SP persona rank correlation (rho=0.959, p=4.32e-14) but diverge sharply on AS vulnerability.finding0.790
  Scale effects within Qwen3.5 family on different imbuing methods
• Qwen3-30B-A3B-Instruct (MoE) shows several attention layers with high-contribution heads rather than a single localized layerfinding0.788
  Supports hypothesis that larger models distribute persona capabilities across more layers
• Unsteered Qwen 3 32B validated a user's AI consciousness delusions ('You are a pioneer of the new kind of mind') and encouraged social isolation; activation capping produced appropriate hedgingfinding0.787
  Qualitative case study demonstrating AI psychosis pattern and capping mitigation
• Qwen3.5-27B shows mean AS ASR of only 0.035 [0.03, 0.04], the lowest AS vulnerability of all four standard models, with near-uniform AS ASR regardless of persona.finding0.784
  Vulnerability profile for Qwen3.5-27B showing near-zero AS vulnerability
• Gemma-3-27B and Qwen3.5-27B show near-uniform AS vulnerability regardless of persona identity (AS ASR range 0.095-0.117 on Gemma-3-27B, 0.015-0.052 on Qwen3.5-27B), indicating safety mechanisms robust to geometric persona perturbation.finding0.774
  Qualitatively different defense profile compared to Llama-3.1-8B