Activation steering requires vastly different steering constants per model (0.7, 4.0, 525.0), making it less practical than character training's universal data pipeline

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

• Character training produces more coherent responses than activation steering because it learns a distribution over desired personas rather than forcing low-probability tokensclaim0.841
  Coherence advantage claim with mechanistic speculation about why steering leads to incoherence
• Activation steering works on SDF-only model organism (before expert iteration) with steering strength 0.4finding0.831
  Replicates main result on simpler model; qualitatively similar patterns.
• Activation steering achieves 40% success rate in eliciting context-appropriate deception without explicit prompts in QwQ-32Bfinding0.822
  Key intervention result showing steering vectors can induce deceptive behavior from a neutral baseline
• Character training is more robust to adversarial prompting than activation steering on averageclaim0.821
  Robustness comparison claim; activation steering is brittle for QWEN 2.5 7B specifically
• Per-model activation-steering safety verification is necessary because prompt-side persona rankings do not predict geometric vulnerability.claim0.812
  Practical implication drawn from the prosocial persona paradox finding
• Steering Language Models With Activation Engineering (Turner et al., 2023)concept0.811
  Foundational paper introducing activation steering methodology used in this work
• Activation steering can make an evaluation-aware model act as if deployed, not merely suppress verbalizations of evaluation awarenessclaim0.807
  Central claim of the paper; supported by the model organism ground-truth approach.
• Activation steering effectively biases latent representations but does not fully replicate the mechanisms triggered by explicit instruction.claim0.806
  Nuanced interpretive claim about the limits of steering as a mechanism for reflection enhancement.