Q8B most often peaks at layer 20, followed by layer 25; G20B most often peaks at layer 15

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

• Peak layer ℓ* median 10, IQR 0.384finding0.817
  Median layer where S(ℓ) peaks, across seeds.
• Middle-to-late layers (39-50) of QwQ-32B show consistently stable and high LAT classification performance across all datasetsfinding0.775
  Layer-wise analysis revealing which network depths best encode strategic deception semantics
• G20B has 7 steerable, 7 natural, 4 intractable generic traits, with more mass at extremes than Q8Bfinding0.769
  G20B's post-training commits more strongly toward and against particular dispositions, leaving fewer in the steerable middle
• Mid-layers (6-15) achieve peak anchoring because semantic structure differentiates while maintaining coherence, forming a Goldilocks zoneclaim0.763
  Interpretation of E3 layer-wise results; motivates targeted UCCT interventions at layers 8-12
• Layer-wise anchoring peaks in a 'Goldilocks zone' between early and late layers.claim0.755
  Qualitative characterization of optimal anchoring depth.
• The case at approximately the 2/3 layer of LLaMA3.1-8B (Layer 24, satisfying Criteria 1 and 2) aligns with prior studies showing the 2/3 layer optimally predicts human brain activity.finding0.753
  Connects this study's results to Schrimpf et al. 2021 and Caucheteux et al. 2022/2023 findings on brain-LLM alignment.
• Gemma-3-4B-it shows three-stage layer trajectory and S(ℓ) peak despite scale differences in dr and ρdfinding0.747
  E3 backbone generalization finding for Gemma; validates pattern across diverse architectures
• Math/code tasks S ≈ -1.65 at layers 8–12finding0.745
  Task-specific peak anchoring score for structured reasoning domains.