训练得到的total_flops是负数

这是训练结束后得到的部分输出(monitor:173) INFO: In worker #1, the system-related metrics are: {'id': 1, 'fl_end_time_minutes': 125.160295, 'total_model_size': 124440576, 'total_flops': -3000, 'total_upload_bytes': 0, 'total_download_bytes': 16470960, 'global_convergence_round': 0, 'local_convergence_round': 0, 'global_convergence_time_minutes': 0, 'local_convergence_time_minutes': 0}。其中total_flops是-3000，请问会是什么原因导致的呢？是因为数据溢出了吗？

No file, test, or reproducible configuration is provided. Start by reproducing the training run that produced monitor:173 and trace how the system-related metric total_flops is calculated. Confirm the cause of -3000 and add a regression check for the corrected metric value.

Written by the indexing model from the issue text.