-
Notifications
You must be signed in to change notification settings - Fork 981
Pull requests: EleutherAI/gpt-neox
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Switch to using Cuda Flash Attn for Alibi
#1183
by haileyschoelkopf
was merged Mar 13, 2024
Loading…
Improve argument validation for Flash-attn + SWA
#1162
by haileyschoelkopf
was merged Mar 2, 2024
Loading…
Enable passing of
--account to srun / SlurmLauncher
#1126
by haileyschoelkopf
was merged Jan 19, 2024
Loading…
Edge-casing for multi-GPU HF-to-NeoX conversion
#1065
by haileyschoelkopf
was merged Nov 1, 2023
Loading…
Update HF conversion docs following reversion of
is_pipe_parallel check
#959
by haileyschoelkopf
was merged Jun 5, 2023
Loading…
Make conversion scripts compatible with Python 3.8
#877
by haileyschoelkopf
was merged Apr 13, 2023
Loading…
No longer ignore
--iteration when passed to train.py
#869
by haileyschoelkopf
was merged Apr 11, 2023
Loading…
Default
use_cache=False in eval harness integration
#774
by haileyschoelkopf
was merged Feb 3, 2023
Loading…
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.