valalalalala

I have not worked with transformers before and implemented checkpoints a little differently than I see in [run_glue_no_trainer.py](https://github.com/huggingface/transformers/blob/6667b0d7bf0a135b9a85d2d21b7eb25ec8ad58cb/examples/pytorch/text-classification/run_glue_no_trainer.py#L590-L594) and implemented it like [so](https://github.com/luckybit4755/diffusers/blob/a452dc9a1b9b280da26daffd8bf51c8d176c1083/examples/train_unconditional.py#L198-L203) In the interests of consistency it might...

Yeah, that's totally fine if you wouldn't mind. Sorry been a bit distracted as well.