Example config for training Diffusion Transformer

[shunk031]

Hi, @coryMosaicML

Thank you for merging the recent PR #155 related to Diffusion Transformer (DiT). I have a couple of questions regarding the use of DiT in this repository:

1. Is there a specific configuration available for running DiT within this repository?
2. Similar to SD2 and SDXL, is it necessary to train DiT across multiple resolutions in a staged manner?

Your guidance on these points would be greatly appreciated.