Error when training textual inversions in A1111

[viewscreenon]

First off, thanks to TheLastBen for making these colabs. A few weeks ago I was using A1111 on colab to train textual inversions (I had to use the "Use cross attention optimizations" setting so it doesn't run out of memory). But after the recent updates I get this error when starting training:

Preparing dataset...
100% 20/20 [00:01<00:00, 13.67it/s]
No saved optimizer exists in checkpoint
0% 0/150 [00:00<?, ?it/s]Traceback (most recent call last):
File "/content/gdrive/MyDrive/sd/stable-diffusion-webui/modules/textual_inversion/textual_inversion.py", line 503, in train_embedding
scaler.scale(loss).backward()
File "/usr/local/lib/python3.9/dist-packages/torch/_tensor.py", line 487, in backward
torch.autograd.backward(
File "/usr/local/lib/python3.9/dist-packages/torch/autograd/init.py", line 200, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
File "/usr/local/lib/python3.9/dist-packages/torch/autograd/function.py", line 274, in apply
return user_fn(self, *args)
File "/usr/local/lib/python3.9/dist-packages/torch/utils/checkpoint.py", line 157, in backward
torch.autograd.backward(outputs_with_grad, args_with_grad)
File "/usr/local/lib/python3.9/dist-packages/torch/autograd/init.py", line 200, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: Expected is_sm80 to be true, but got false. (Could this error message be improved? If so, please report an enhancement request to PyTorch.)

[TheLastBen]

I removed "--xformers" because it caused the error nansexception when using high resolutions, you can add it back to fix this issue

[viewscreenon]

Thank you, it worked.