- Rating
- 1
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper proposes Bidirectional Circuit Discovery (BiCD), an automated circuit discovery method for mechanistic interpretability. BiCD operates at the component node level under zero ablation and a correct-answer cross-entropy (CE) constraint, to find a small subcircuit capable of performing fine-tuned arithmetic tasks on GPT-2 small and medium.
Critical strengths
Combining dynamic gradient rescoring with a forward-then-backward search trajectory helps account for context-dependent component interactions while reducing the number of exact evaluation trials compared to static rankings.
The writing is clear.
Critical weaknesses
Most importantly, the proposed combination of zero ablation and node-level pruning reverts to an early formulation whose limitations are already well-documented in the mechanistic interpretability literature—precisely the limitations that motivated the community's shift toward interchange (corrupted, you mean) patching and edge-level pruning. First, zero ablation severely ruins the activation space and pushes representations out-of-distribution; forcing a component's output to zero does not correspond to rendering a neuron or component functionally "inactive" or neutral. Second, prior foundational works (such as Path Patching and IOI circuit analysis) have long established that complex, fine-grained edge paths are required to characterize true algorithmic mechanisms, whereas coarse node pruning (especially while retaining all 12 MLP layers unconditionally) obscures the underlying causal routing.
Building on this paradigm shift, the authors discard standard evaluation frameworks in favor of a standalone zero-ablation accuracy metric, claiming that subcircuits found by activation patching methods "often have limited ability to perform the task independently" (Section 1 and Appendix F.2). In reality, under standard interchange/corrupt ablation evaluations (e.g., EAP and EAP-IG on the canonical IOI task), existing methods reliably recover over 90% of the model's task performance and faithfulness. Evaluating methods or objectives designed for distribution-preserving patching under harsh zero-ablation conditions—and using a fine-tuned checkpoint that suffers catastrophic forgetting on standard benchmarks—does not substantiate the claim that existing methods fail.
Questions for authors
-
How do the authors justify that setting a component's output to zero represents the component being "inactive"? If a component is genuinely inactive on a given task, would its natural output ever be an all-zero matrix (particularly in the presence of biases and non-zero baseline activations), and what strict conditions would its input have to satisfy for that to occur?
-
Consider two attention heads and an MLP . Suppose Mechanism 1 involves reading directly and independently from both and ( and ), whereas Mechanism 2 involves passing information to , which then passes it to (, with no direct functional path ). Since and must all be retained as nodes in both cases, how can the proposed node-level circuit representation distinguish between these two fundamentally different mechanisms?