How to prepare dataset for pretraining after https://github.com/NVIDIA-NeMo/NeMo/pull/14192

[ooooo-create]

I see the document for custom Pretrain dataset in https://docs.nvidia.com/nemo-framework/user-guide/latest/data/pretrain_data.html . It will use scripts/nlp_language_modeling/preprocess_data_for_megatron.py which is removed in #14192. I want to know there are others ways to do that. Thanks!