RuntimeError: cublas runtime error : the GPU program failed to execute at pytorch/aten/src/THC/THCBlas.cu:249

nlp
1

The above error occurred when I used fairseq to train the language model. I don’t know what to do. I used torch0.4.1

2

Could you update PyTorch to the latest stable version (1.6) and rerun the code again, please?