Tutorial BUG

[nrailg]

Describe the bug

1. GPTDatasetConfig got unexpected keyword mmap_bin_files (can be solved if I install main of Megatron LM instead of megatron-core-r0.5)
2. GPTDatasetConfig got unexpected keyword is_build_on_rank
3. merge file and vocab file seem to be useless, since AutoTokenizer downloads gpt2 repo anyway.

Steps/Code to reproduce bug

follow https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/nlp/nemo_megatron/gpt/gpt_training.html

Environment overview (please complete the following information)

Cuda 12 Torch 2.1 Megatron-LM 9de386d08770d7296263a590171ace4ae45348ad NeMo e64b2227b4759665187d061784927d2f9d0868b3

[github-actions[bot]]

This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.

[github-actions[bot]]

This issue was closed because it has been inactive for 7 days since being marked as stale.