method
active
method:attribution-methodsAttribution methods
Measure causal importance of upstream variables.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- Gradient-based attribution methodsrelated_toUsed by APD; poor approximations of causal importance.
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.
- The task of attributing model behaviors to specific training datapoints.
- Correlating attribution vectors (feature activation × logit weight of next token) across model pairs to measure functional universality
- Gradient-based method to estimate the effect of zeroing a feature on a specific logit difference.
- Attribution methods are necessary for causal explanations but are often difficult to interpretclaim0.809Open problem in attribution.
- Dot product of summed decoder vector with unembedding vector, following Hanna et al.
- Gradient-based technique using SAE features to estimate causal effects on completions; used to corroborate NLA findings.
- Baseline method against which probe-based ranking is compared; more computationally expensive.
- Core method decomposing attention scores as bilinear sums of query-key feature pairs.