method
active
method:activation-oracles-aoActivation Oracles (AO)
Supervised method training models to answer questions about activations; NLAs differ by being unsupervised.
Neighborhood — ranked by edge-count
Methods (1)
method
- Natural Language Autoencoders (NLAs)associated_withextendsCore unsupervised method for generating natural language explanations of LLM activations through a verbalizer-reconstructor pair trained with RL.
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.
- Framework training LLMs to answer questions about externally-provided activation vectors
- Internal representations of the model on which probes operate; the method uses activations to rank datapoints.
- Intervention method that adds a learned direction vector to residual stream activations to steer model behavior
- Large activation magnitudes in the residual stream that Queipo-de Llano et al. link to causing compression behavior and stages of inference
- Causal intervention technique: edit NLA explanation, reconstruct via AR, use difference as steering vector to manipulate model behavior.
- Clamping activations along the Assistant Axis to remain above a minimum threshold (25th percentile), introduced as a stabilization method
- The view of AI as a question-answer system optimized for correctness, often inherited from supervised learning.
- Standard method in mechanistic interpretability that intervenes on activations; VPD flips this paradigm by patching parameters.