Scaling Diffusion Transformers Efficiently via $\mu$P | Read Paper on Bytez