[Llama 3.1] Add LLaMA 3.1 8B parameter weights to Google Drive folder

[davidjurado]

I'm currently experimenting with an integration with MLCube, and it would be ideal to have the LLaMA 3.1 8B parameter weights added to the existing Google Drive folder that currently holds the 405B model weights. The main goal is to use the 8B model for debugging and testing purposes, as it will allow for faster iteration with this smaller variant. To maintain the structure of the folder, we could add a new folder called '8B' next to the existing '405B' folder under MLCommons Llama 3.1/training/nemo-formatted-hf-checkpoint

[hiwotadese]

@Elnifio can you take a look at this?

[nathanwasson]

The checkpoint is now in the requested location in a "8b" directory.

[hiwotadese]

@davidjurado can you try this?