finding
active
finding:steering-example-swapping-the-operator-in-a-math-problem-on-r1Steering example: swapping the operator in a math problem on R1
Demonstration of feature steering on R1.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- Modifying model behavior by clamping SAE feature activations to specific values during forward pass.
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.
- Demonstration of steering reasoning length.
- First steering insight specific to R1.
- Core mechanism statement.
- Steering vectors from µ(0→2) slightly outperform µ(1→2) for instruction discovery across datasets and modelsfinding0.771Shows that contrasting No Reflection with Triggered Reflection provides a stronger signal than Intrinsic vs Triggered.
- Steering vector from mean activation difference between contrasting datasets.
- Explanation of brittle steering.
- Foundational paper introducing activation steering methodology used in this work