Is there any example with recent version of Megatron-LM?

[cryoco]

trafficstars

The examples showed here or here is based on versions about half a year ago. Is there any examples aligned with recent Megatron? Or, is there still relatively obvious optimization with deepspeed to Megatron with pipeline parallelization now?

[zhaowei-wang-nlp]

Hi but the second page says that it is updated on March 21, 2022. I am new to Deepspeed and learning it. I don't know whether it is true.