🐛 Describe the bug

I'm seeing the following error when running with dynamo

RuntimeError: Failed running call_function aten.nll_loss_backward(*(FakeTensor(FakeTensor(..., device='meta', size=(1,)), cpu), FakeTensor(FakeTensor(..., device='meta', size=(3,)), cpu)), **{'target': FakeTensor(FakeTensor(..., device='meta', size=(3,), dtype=torch.int64), cpu), 'weight': None, 'reduction': 0, 'ignore_index': 10, 'total_weight': FakeTensor(FakeTensor(..., device='meta', size=()), cpu)}):
Index tensor must have the same number of dimensions as self tensor

repro:

import torch
import torch._dynamo as dynamo


@dynamo.optimize("eager")
def forward(grad_output, input, target, total_weight):
    return torch.ops.aten.nll_loss_backward(grad_output,
                                            input,
                                            target=target,
                                            weight=None,
                                            reduction=0,
                                            ignore_index=10,
                                            total_weight=total_weight)


forward(torch.randn(1), torch.randn(3), torch.tensor([2, 3, 0]),
        torch.tensor(3.))

Full error log: https://gist.github.com/silvasean/5dd4b1667c9e451d12dac84e8a473173

Versions

Collecting environment information...
PyTorch version: 1.14.0.dev20221114+cpu
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: Debian GNU/Linux rodete (x86_64)
GCC version: (Debian 12.2.0-3) 12.2.0
Clang version: 14.0.6-2
CMake version: version 3.25.0
Libc version: glibc-2.35

Python version: 3.10.8 (main, Nov 3 2022, 15:17:13) [GCC 12.2.0] (64-bit runtime)
Python platform: Linux-5.18.16-1rodete4-amd64-x86_64-with-glibc2.35
Is CUDA available: False
CUDA runtime version: No CUDA
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

Versions of relevant libraries:
[pip3] numpy==1.23.4
[pip3] torch==1.14.0.dev20221114+cpu
[pip3] torchvision==0.15.0.dev20221114+cpu
[conda] Could not collect

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @kadeng @chauhang @amjames @Lucaskabela @jataylo @chenyang78 @aakhundov @mlazos @soumith @yanboliang @anijain2305 @chunyuan-w @Xia-Weiwen @desertfire