"megatron/training/theoretical_memory_usage.py" did not exist on "3aca141586a4b8cdc983c3ecf5f7baf60506c7f8"
-
Mirza Halilčević authored
Since opset version 18, the Split operator allows splitting into unevenly sized outputs when the split input is not present, and a num_outputs attribute has been introduced.
03afba77
