Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Info on Deberta-v2-xlarge training infra #125

Open
karthickgopalswamy opened this issue Mar 22, 2023 · 0 comments
Open

Info on Deberta-v2-xlarge training infra #125

karthickgopalswamy opened this issue Mar 22, 2023 · 0 comments

Comments

@karthickgopalswamy
Copy link

The paper talks about DeBERTa-large, base and DeBERTa1.5B model on V100 GPU. How is the DeBERTa-v2-xlarge trained? is the settings for the xlarge model same as used for large model in the paper? With DeBERTa-v2-xlarge having 900M parameters is any tensor parallelism used for training?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant