Wellbeing probe: peak Cohen's d=3.34 (layer 16), p=7.21×10⁻¹³ in LLaMA-3.2-3B

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

• Interest probe: peak Cohen's d=1.67 (layer 14), p=9.45×10⁻⁶ in LLaMA-3.2-3Bfinding0.885
  Probe validation result confirming interest direction captures meaningful structure
• Qwen 2.5 7B wellbeing probe: peak Cohen's d=3.5finding0.883
  Strongest cross-family probe; explains clearer introspection in Qwen than Gemma
• Gemma 3 4B wellbeing probe: peak Cohen's d=1.8finding0.880
  Weaker cross-family probe; explains weaker introspection in Gemma
• Impulsivity probe: peak Cohen's d=3.60 (layer 13), p=3.58×10⁻¹³ in LLaMA-3.2-3Bfinding0.870
  Strongest probe validation result; highest Cohen's d among the four concepts
• LLaMA-3.1-8B-Instruct wellbeing introspection: ρ=0.93, isotonic R²=0.90 (LMM probe slope p<10⁻¹⁰)finding0.827
  Near-ceiling introspective performance for wellbeing concept in 8B model; nearly deterministic probe-report relationship
• Wellbeing probe-score drift across turns significant at all three LLaMA scales (slopes=0.006, 0.005, 0.013 for 1B, 3B, 8B; all p<10⁻¹⁰); drift magnitude increases with scalefinding0.827
  Internal-state drift generalizes across scales; normalized drift also increases significantly with log(model size)
• Wellbeing concept: Spearman ρ=0.68, isotonic R²=0.48 in LLaMA-3.2-3B (n=400, p<10⁻²⁶)finding0.793
  Second-strongest pooled introspective coupling in primary model
• Llama-3.1-405B rates representative coin sequences 5.38 vs. 3.57 for non-representative, Cohen's d=5.15, p<10^-6finding0.780
  Validates Assumption D.6 that base models assign higher typicality ratings to representative (diverse) sequences