method
active
method:predictor-agent-uplift-evaluationPredictor agent uplift evaluation
Evaluation of tool uplift over transcript-only baseline using accuracy and AUROC.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Auditing gamesusesMethodology measuring interpretability tool uplift for an agent uncovering hidden behaviors.
Methods (2)
method
- Core unsupervised method for generating natural language explanations of LLM activations through a verbalizer-reconstructor pair trained with RL.
- Supervised method training models to answer questions about activations; NLAs differ by being unsupervised.
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.
- Agent that plans, implements, and revises behavioral evaluations using Claude Code.
- Production assessment.
- Evaluation agent implemented Meinke et al. scheming evaluation from blog post and abstract onlyfinding0.701Fully designed and implemented.
- Central interpretive claim from statistical analysis
- Demonstrates practical utility: NLAs enable root-cause discovery without access to misaligned model's training data.
- Prediction orthogonality thesis.
- Demonstration that model-level priors (not parameter-level knowledge) suffice for immediate transfer
- Conclusion.