[2025-12-30T06:43:27.005] error: Prolog hung on node gpub073
W1230 06:50:52.141555 1118861 torch/_dynamo/convert_frame.py:1016] [5/8] torch._dynamo hit config.recompile_limit (8)
W1230 06:50:52.141555 1118861 torch/_dynamo/convert_frame.py:1016] [5/8]    function: '_base_sub' (/u/yurenh2/.local/lib/python3.9/site-packages/snntorch/_neurons/leaky.py:242)
W1230 06:50:52.141555 1118861 torch/_dynamo/convert_frame.py:1016] [5/8]    last reason: 5/7: tensor 'input_' size mismatch at index 2. expected 128, actual 256
W1230 06:50:52.141555 1118861 torch/_dynamo/convert_frame.py:1016] [5/8] To log all recompilation reasons, use TORCH_LOGS="recompiles".
W1230 06:50:52.141555 1118861 torch/_dynamo/convert_frame.py:1016] [5/8] To diagnose recompilation issues, see https://pytorch.org/docs/main/torch.compiler_troubleshooting.html.
Traceback (most recent call last):
  File "/projects/bfqt/users/yurenh2/ml-projects/snn-training/files/experiments/lyapunov_speedup_benchmark.py", line 638, in <module>
    run_scaling_test(args.device)
  File "/projects/bfqt/users/yurenh2/ml-projects/snn-training/files/experiments/lyapunov_speedup_benchmark.py", line 596, in run_scaling_test
    model_base = BaselineSNN(**cfg).to(device)
TypeError: __init__() got an unexpected keyword argument 'batch_size'