-
Notifications
You must be signed in to change notification settings - Fork 970
Issues: EleutherAI/gpt-neox
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[BUG?] Higher "gradient_accumulation_steps" still increases memory usage a lot
bug
Something isn't working
#1123
opened Jan 15, 2024 by
exnx
Adding AxoNN's 3D tensor parallelism [WIP]
feature request
New feature or request
#1086
opened Nov 28, 2023 by
siddharth9820
•
Draft
1 of 3 tasks
Port DeepSpeed Ulysses
feature request
New feature or request
#1078
opened Nov 12, 2023 by
Quentin-Anthony
The plot got from muP coord_check seems not horizontal, which may indicates there exits a bug in the muP implementation?
bug
Something isn't working
#956
opened May 28, 2023 by
BaoYu0721
Adds a script to convert NeoX 2.0 checkpoints to DeepSpeed's universal checkpoint format
merge-queue
This PR is next on the queue to merge
#836
opened Mar 14, 2023 by
dashstander
Loading…
Add support for sequence parallelism
feature request
New feature or request
help wanted
This issue needs assistance
#812
opened Mar 7, 2023 by
Quentin-Anthony
ProTip!
Add no:assignee to see everything that’s not assigned.