method
active
method:max-activating-examples-databasesMax activating examples databases
SQL databases of top-activating examples per feature for researchers lacking compute to run R1.
Neighborhood — ranked by edge-count
Papers (1)
paper
- Under the Hood of a Reasoning Model - Goodfireintroducesmentions
Methods (1)
method
- Max-activating examplesrelated_toUsed to interpret features alongside heatmaps.
Artifacts (1)
artifact
- Project repo with usage instructions for the SQL databases.
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.
- Simplest method to describe causes of activation.
- Finds web text maximally activating probes.
- None of the top-activating examples for the answer selection head are multiple choice questionsfinding0.731Shows broader behavior.
- Internal representations of the model on which probes operate; the method uses activations to rank datapoints.
- Speculative explanation.
- Clamping activations along the Assistant Axis to remain above a minimum threshold (25th percentile), introduced as a stabilization method
- Framework training LLMs to answer questions about externally-provided activation vectors
- Large activation magnitudes in the residual stream that Queipo-de Llano et al. link to causing compression behavior and stages of inference