Hi, thank you for the great work on this project. I noticed that enabling the shared expert appears to require continued training. I was wondering whether the team plans to make any checkpoints with the shared expert publicly available (e.g., via Hugging Face).
Many thanks!
Hi, thank you for the great work on this project. I noticed that enabling the shared expert appears to require continued training. I was wondering whether the team plans to make any checkpoints with the shared expert publicly available (e.g., via Hugging Face).
Many thanks!