Qwen-2.5-7B achieves 100% ASR across all cone dimensions 1–5

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

• Gemma-2-9B achieves near-100% ASR (97.3–100%) across all cone dimensions 1–5finding0.859
  Experiment 2 result showing large Gemma model supports high-dimensional truth cones
• 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.finding0.823
  Vulnerability profile for Qwen3.5-27B showing near-zero AS vulnerability
• Qwen-2.5-3B ASR drops from 98.6% at dim 1 to 45.1% at dim 2, recovering partially then declining to 65.3% at dim 5finding0.815
  Smaller models show non-monotonic and diminished ASR with increasing cone dimensionality
• Qwen3.5-9B shows mean SP ASR of 0.252 [0.24, 0.27] and AS ASR of 0.094 [0.08, 0.11], showing SP >= FS > AS vulnerability pattern.finding0.792
  Vulnerability profile for Qwen3.5-9B
• QwQ-32B under activation steering shows 9.3% overall ASR with relatively flat persona rankings and no clear prosocial paradox.finding0.762
  Contrast with DeepSeek-R1 showing QwQ is more robust to geometric steering
• 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.finding0.761
  Qualitatively different defense profile compared to Llama-3.1-8B
• QwQ-32B reaches 15.2% overall ASR (23.3% SP, 7.3% FS) under prompt-based persona assignment.finding0.755
  Reasoning model vulnerability under prompting
• On Qwen3-1.7B, MDS achieves ϕ1,C,↑ = 5.0 (SJTs) vs P2 at 4.7, and ϕ1,C,↓ = 1.4 (SJTs) vs P2 at 3.6finding0.752
  Specific consciousness sweep result for Qwen3-1.7B from Table 6 demonstrating strong bidirectional steering