Idea: Speed up ACDC by batching edge-checks. The intuition is that most circuits will have short paths through the transformer, because Residual Networks Behave Like Ensembles of Relatively Shallow Networks (https://arxiv.org/pdf/1605.06431.pdf). Most edges won't be in most circuits. Therefore, if you're checking KL of random-resampling edges e1 and e2, there's only a small chance that e1 and e2 interact with each other in a way important for the circuit you're trying to find. So statistically, you can check eg e1, ... e100 in a batch, and maybe ablate 95 edges all at once (due to their individual KLs being low). 

If true, this predicts that given a threshold T, and for most circuit subgraphs H of the full network graph G, and for the vast majority of e1, e2 in H: 

KL(G || H \ {e2} ) - KL(G || H) < T 
iff
KL(G || H \ {e2, e1} ) - KL(G || H \ {e1}) < T 

(That is, e1's inclusion doesn't change your pruning decision on e2)

Neel Nanda suggests that it's particularly natural to batch edges in the same layer.