Question about sft with deepspeed

Hello , I face a problem when training mistral model in a sft way with deepspeed zero3 config, here is the error information:

Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: The size of tensor a (0) must match the size of tensor b (14336) at non-singleton dimension 1

Which problem occurs in loss computation process and do you have any suggestions about this problem？