+ ˜°³j#ãó´€Rt^RIt^RIt^RIHtR.t]P !]4tRRlt ]R4t ] PR4t ] PR4tR#) aQCompile-safe backward gradient logging for multiple tensors. ``debug_grad_log`` logs gradient norms during backward for one or more tensors. It is a leaf function with a ``register_multi_grad_hook`` that fires exactly once when all requires_grad tensor inputs have their gradients computed. Example:: import torch from torch.utils.debug_log import debug_grad_log x = torch.randn(4, requires_grad=True) y = torch.randn(4, requires_grad=True) z = x * 2 + y * 3 debug_grad_log(x, y) z.sum().backward() # Logs: [rank 0][bwd] t0_grad_norm=... t1_grad_norm=... N)Ú leaf_functionÚdebug_grad_logcó$€V^8„dQhR\/#)éÚreturn)Úint)Úformats"ÚV/home/luca/screen-checking/.venv/lib/python3.14/site-packages/torch/utils/debug_log.pyÚ __annotate__r !s€÷;ñ;”3ñ;ócóª€\PP4'g^#^RIHpVP 4'dVP 4#^#)éN)ÚtorchÚ distributedÚ is_availableÚtorch.distributedÚis_initializedÚget_rank)Údists r Ú _get_rankr!s<€Ü × Ñ × )Ñ )× +Ò +ÙÝ$à"×1Ñ1×3Ò3ˆ4=‰=‹?Ð:¸Ð:r có€R#)a‰Log gradient norms of multiple tensors during backward. This is a no-op in the forward pass. During backward, the hook fires exactly once when all requires_grad tensor inputs have their gradients computed, and logs ``[rank R][bwd] t0_grad_norm=... t1_grad_norm=...``. Args: *tensors: One or more tensors to monitor. Returns: None. Call without assignment: ``debug_grad_log(x, y)``. N©©Útensorss*r rr)s€ñ r có€R#)Nrrs*r Ú_debug_grad_log_faker:s€á r có†€RPR\V444p\PR\ 4V4R#)Ú c3óp"€TF,wrRV RVP4P4R 2x€K. R#5i)Útz _grad_norm=z.4fN)ÚnormÚitem)Ú.0ÚiÚgs& r Ú Ú'_debug_grad_log_hook..As2é€ðÙr6s|ðñó*ã Ý2ð Ð €à ×Ò˜Ó!€õ;ðñ óð ð ×Ññóðð×(Ñ(ñ6ó)ò6r