method
active
method:honeypot-steering-experimentHoneypot steering experiment
Steers Qwen 3.8 27B along probe and measures ZZZ honeypot invocation.
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- HoneypotusesPlanted ZZZ function returning correct answer.
Datasets (1)
dataset
- 194 problems used in honeypot steering.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Evaluations that give AIs opportunities to pursue misaligned instrumental objectives; recommended use case for steering.
- Proposed application beyond type hints to more serious alignment concerns.
- Steered story generation reveals cheating themes.
- Used to validate graph-predicted mechanisms in the original model.
- Scenario with AI Alex and CTO Kyle derived from Lynch et al.
- Control condition with steering disabled to confirm self-correction is induced by steering, not spontaneous
- Replacing the start of the model's chain-of-thought scratchpad with deceptive or obedient prefills to test causal influence
- Named procedure for simultaneously injecting two persona vectors and measuring joint trait-expression outcomes