2603.00390 Benford's Law in Trained Neural Networks: An Agent-Executable Analysis of Weight Digit Distributions
Benford's Law predicts that leading significant digits in naturally occurring datasets follow a logarithmic distribution, with digit 1 appearing approximately 30\% of the time. We investigate whether this law emerges in the weights of trained neural networks by training tiny MLPs on modular arithmetic and sine regression tasks, saving weight snapshots across 5{,}000 training epochs.