method
active
method:model-surgeryModel Surgery
Edits MLP weights for all layers to modify model behavior; used by Abdelnabi & Salem to decrease verbalized evaluation awareness.
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.
- A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
- Technique to measure representational compatibility by integrating intermediate representations of one model into another
- Technique for modifying model knowledge or behavior via targeted interventions, e.g., ROME by Meng et al.
- Comparing models using log-evidence approximated by free energy.
- LLM behavior of generating falsehoods; the multi-dimensional truth subspace raises new risks for subtle manipulation
- Large pretrained models used as backbones across tasks; their universality motivates the convergence hypothesis
- Primary substrate for manifold steering experiments; demonstrates method on reasoning and in-context tasks.