Neuron Level Analysis of Large Language Model in Legal Domain Reasoning
For researchers studying LLM interpretability and domain-specific reasoning, this work provides insights into neuron specialization and overlap in legal tasks, though it is incremental as it applies existing attribution methods to a new domain.
The study performs neuron-level analysis of legal-domain reasoning in LLMs, identifying influential neurons whose suppression collapses task accuracy while random suppression does not, and revealing task-specific and cross-task neurons across seven models.
We presented a neuron-level analysis of legal-domain reasoning in LLMs, comparing it with other applied domain tasks across seven open-weight models. Using neuron attribution scores to rank and suppress influential neurons, we confirmed that suppressing the identified neurons collapses accuracy on the target task, whereas suppressing the same number of random neurons does not. We further found a small subset of neurons influential across all seven tasks; once these are removed, suppressing the remaining neurons degrades only the task they were identified from, revealing genuinely task-specific neurons in every model studied. Within the legal domain, the three benchmarks exhibit relatively high neuron overlap and tend to be affected jointly, suggesting of legal components neurons that span jurisdictions. The distribution of identified neurons in our experiments suggests that the hypothesis that influential neurons are concentrated in middle MLP layers may depend on the input format and content, rather than being a universal phenomenon.