How to properly reference the use of the ESM-3 open version in papers and GitHub repositories?

[rokieplayer20]

Hello,

Thank you so much for releasing the outstanding ESM-3 open model to the public in non-comercial.

I have developed an AI model using protein embeddings extracted from the ESM-3 open version. I plan to share the protein embeddings and the trained model parameters in both an academic paper and on GitHub.

I have read the content available on the ESM-3 GitHub page. However, since it is not feasible to include the full details of the "Cambrian Non-Commercial License Agreement," I am asking how to reference it in the simplest and most concise way possible.

Here are some questions I would like to ask:

1. I have extracted protein embeddings from publicly available and widely recognized datasets (e.g., BindingDB).
   Am I allowed to share these embeddings in the GitHub repository or Google Drive linked to my paper?
   If sharing is permitted, could you provide guidelines or examples of the preferred format for sharing them? I am happy to follow the recommended practices.

2. In my paper and GitHub page, I plan to acknowledge the use of ESM-3 open and include a note about complying with the policies outlined in (https://www.evolutionaryscale.ai/policies/cambrian-non-commercial-license-agreement).

Would this be sufficient to freely share my code and the trained model parameters?
If additional steps are required, I would greatly appreciate it if you could provide further guidelines or example formats.

The ESM series is making significant contributions to the advancement of human health and academic research.

I deeply appreciate your hard work and dedication. Thank you!