method
active
method:gradient-based-attribution-methodsGradient-based attribution methods
Used by APD; poor approximations of causal importance.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Decomposes parameters into faithful, minimal, simple components.
Methods (2)
method
- Gradient-based data attributionrelated_toBaseline method against which probe-based ranking is compared; more computationally expensive.
- Attribution methodsrelated_toMeasure causal importance of upstream variables.
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.
- Some gradient-based attribution methods give attributions independent of model and data-generating processfinding0.899Adebayo et al. sanity checks.
- Backprop of logit-minus-mean-logit loss to surface causal intervention candidates, summed across token positions.
- Optimization technique that computes weight changes by following the gradient of an error function; contrasted with evolutionary stochastic search.
- Used for updating hidden state expectations; provides dynamical process theory testable against neuronal data
- Motivation.
- The property that qualities vary slowly, subtly, gradually across the extent of each living thing; gradients arise as natural responses to changing circumstances and create field-like character that points toward and establishes centers
- Linear classifier approach applied to model activations to identify which training datapoints caused undesired behaviors in post-training.
- Frequent-task updates overwrite rare-task features.