method
active
method:full-replacement-modelFull replacement model
Substitutes all MLP outputs with CLT reconstructions.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- MLP replacement with sparse features underlying the original attribution graphs.
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.
- CLT plus error nodes and original attention patterns, local to a prompt.
- When a single conversation involves different models, e.g., routing between standard and reasoning models
- A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
- Metrics of how thoroughly features explain computation.
- Large pretrained models used as backbones across tasks; their universality motivates the convergence hypothesis
- Later layers correct single-layer steering interventions.
- Change point segmentation, kernel Dirichlet pooling, CV tuning.
- Comparing models using log-evidence approximated by free energy.