Llama-3.3-70B corrected response scores 75/100 rather than 100 due to residual steering effects (Snell's law reference)

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

• Fine-tuning Llama-3.1-8B on self-correction examples increases multi-attempt rate proportionally with training data ratiofinding0.835
  Shows behavioral pattern of self-correction is trainable in smaller models
• Llama-3.3-70B shows multi-attempt rate of 7.4% vs. ≤1.2% for all other models testedfinding0.832
  Supporting finding showing ESR is driven by both higher multi-attempt rates and comparable improvement rates
• Meta-prompting increases Llama-3.3-70B multi-attempt rate 4.3× (from 7.4% to 31.7%)finding0.808
  Demonstrates ESR can be deliberately enhanced through prompting in the largest model
• Llama 3.1 405B shows 14% compliance gap in minimal helpful-only replication but smaller Llama and Mistral models show no gapfinding0.806
  Replication across open-weight models supports scale-emergence finding
• Llama-3.3-70B exhibits internal consistency-checking mechanisms that operate during inferenceclaim0.806
  Central interpretive claim of the paper supported by causal ablation and activation evidence
• Steering Llama-3.1 8B along the circular representation manifold produces outputs that follow the natural circle of the behavior manifold, cleanly shifting probability mass from Monday through successive days.finding0.803
  Core empirical result demonstrating that manifold steering produces on-target, behavior-aligned outputs.
• Correlation between layer-wise scores and task accuracy ρ = −0.73 (p < 0.001) on LLaMAfinding0.803
  Core E3 finding validating S as a predictor of anchoring effectiveness
• Llama 3.3 70B is the most likely to take on a non-Assistant persona when steered, with even split between human and nonhuman portrayalsfinding0.802
  Model-specific difference in persona susceptibility