Megatron-LM
Megatron-LM copied to clipboard
Build dataset for all GPUs with tp_rank=0 and pp_rank=0 or -1 in multi-machine training.
Not just for the GPU with global_rank=0, following the code below. https://github.com/NVIDIA/Megatron-LM/blob/28118fcdc22e42621776a021af568ae39c198418/pretrain_gpt.py#L257-L260 https://github.com/NVIDIA/Megatron-LM/blob/28118fcdc22e42621776a021af568ae39c198418/pretrain_gpt.py#L306-L311
Marking as stale. No activity in 60 days.
This PR was closed because it has been inactive for 7 days since being marked as stale.