Post-training steers models toward a particular region of persona space but only loosely tethers them to it, motivating work on training and steering strategies that more deeply anchor models to a coherent persona

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

• How does different post-training data shift a model's position along persona dimensions?question0.846
  Future work direction: using persona space to study effects of training data on model character
• Post-training is key to eliciting strong introspective awareness; base pretrained models do not show above-chance detectionclaim0.821
  Finding that base models have high false positives and no net positive performance.
• Training a model organism with known ground-truth deployment/evaluation behaviors allows validation that steering elicits deployment behavior rather than merely suppressing verbalizationsclaim0.815
  Methodological claim distinguishing this paper from prior work on verbalization suppression.
• Post-training strategies can strongly influence performance on introspective tasksclaim0.813
  Assertion about the role of post-training in eliciting introspection.
• Post-training is key to eliciting introspective awarenessfinding0.803
  Base pretrained models show high false positive rates and achieve no net task performance on concept injection detection; post-training essential for introspection.
• Post-training influences introspective capability expressionclaim0.798
  Different post-training strategies substantially influence introspection task performance; 'helpful-only' variants show higher false positives but some achieve strong net performance.
• The Assistant Axis in instruct models mainly inherits from pre-existing helpful and harmless human personas in base models, later acquiring additional associations (such as being an AI) during post-trainingclaim0.793
  Key mechanistic claim about the developmental origin of the Assistant persona
• Steering vector extracted from final post-expert-iteration model also successfully elicits deployment behaviorfinding0.783
  Replicates main result using in-distribution steering vector; addresses concern about pre-trained vector validity.