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.

Sign up for GitHub

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

Jump to bottom

For 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work. #35

Open
joytianya opened this issue Apr 19, 2023 · 2 comments
Open

For 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work. #35

joytianya opened this issue Apr 19, 2023 · 2 comments

Comments

@joytianya
Copy link

joytianya commented Apr 19, 2023 •
edited

For 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work.
@joytianya joytianya changed the title Can the for 30B LLama model, server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work. for 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work. Apr 19, 2023
@joytianya joytianya changed the title for 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work. for 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-256 (128g)? I tried 1, 64,4 and 1,32,8 but they don't seem to work. Jun 16, 2023
@joytianya joytianya changed the title for 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-256 (128g)? I tried 1, 64,4 and 1,32,8 but they don't seem to work. For 30B LLama model, can server be supported by configuring mesh_dims on tpu v3-8 (128g)? I tried 8,1 and 4,1 but they don't seem to work. Jun 16, 2023
@versae
Copy link

versae commented Sep 3, 2023

Any luck training the 30B on a single TPU v3-8 so far? Does it even fit? The 7B needs 84GB of VRAM, so I would expect the 30B to need at least 4 times that.

@joytianya
Copy link
Author

Currently, there is still no work.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@versae @joytianya