Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis

Open in new window