-
Notifications
You must be signed in to change notification settings - Fork 982
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
Dataset downloads <number of GPUs> times when running deepspeed train.py
#37
by sdtblck
was closed Jan 5, 2021
Write dataset class that tokenizes on the fly
feature request
New feature or request
#40
by sdtblck
was closed Feb 15, 2021
Fix tfrecord dataset to load less files into memory
bug
Something isn't working
#41
by sdtblck
was closed Mar 2, 2021
Add Deepspeed Transformer Kernel
feature request
New feature or request
good first issue
Good for newcomers
#43
by sdtblck
was closed Jan 10, 2021
Ensure learning rate scheduler is functioning correctly
bug
Something isn't working
documentation
Improvements or additions to documentation
#44
by sdtblck
was closed Jan 23, 2021
Implement Pipeline Parallelism
feature request
New feature or request
#45
by sdtblck
was closed Jan 14, 2021
Implement Generation / Eval with deepspeed model engine
feature request
New feature or request
#58
by sdtblck
was closed Feb 15, 2021
Ensure Checkpoint Saving / Loading works correctly
bug
Something isn't working
#151
by sdtblck
was closed Mar 6, 2021
Ensure Sampling works correctly
experiments
Experiments we wish to perform on the codebase
feature request
New feature or request
#152
by sdtblck
was closed Apr 22, 2021
Integrate HuggingFace Tokenizers
feature request
New feature or request
good first issue
Good for newcomers
#190
by sdtblck
was closed Apr 7, 2021
Number of GPUs not automatically detected on a single node instance
#194
by sdtblck
was closed Apr 5, 2021
Why doesn't 1-bit adam improve pipeline parallel speed?
bug
Something isn't working
#206
by sdtblck
was closed Dec 30, 2021
Can we get sparse attention working with A100s / CUDA 11?
feature request
New feature or request
#207
by sdtblck
was closed Apr 9, 2021
Code Cleanup
documentation
Improvements or additions to documentation
#208
by sdtblck
was closed Apr 30, 2021
3 tasks done
Add Progressive Growing of Batch Size
feature request
New feature or request
#215
by sdtblck
was closed Sep 29, 2023
Add Unit Tests / Style Tests
documentation
Improvements or additions to documentation
feature request
New feature or request
good first issue
Good for newcomers
#221
by sdtblck
was closed Apr 28, 2021
4 tasks
Fix Fused Kernel compiler getting stuck
bug
Something isn't working
#232
by sdtblck
was closed Jun 23, 2021
Timer logging innacurate if pp=0
bug
Something isn't working
#238
by sdtblck
was closed Apr 30, 2021
Get rid of codepath where pp = 0
feature request
New feature or request
#243
by sdtblck
was closed Apr 30, 2021
Previous Next
ProTip!
Follow long discussions with comments:>50.