-
Notifications
You must be signed in to change notification settings - Fork 2.8k
Pull requests: NVIDIA/NeMo
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Fix GPT HF Exporter dtype and head_dim
#12792
opened Mar 26, 2025 by
kevalmorabia97
Loading…
2 of 8 tasks
Add BERT/Qwen2.5 Unit test and Refactor all GHA Conversion Tests
CI
Run CICD
#12785
opened Mar 26, 2025 by
suiyoubi
Loading…
8 tasks
change runner to use hopper for testing sequence packing
CI
Run CICD
#12782
opened Mar 26, 2025 by
yashaswikarnati
Loading…
8 tasks
Adjust rank ordering
NLP
Run CICD
skip-docs
#12781
opened Mar 26, 2025 by
ryantwolf
Loading…
3 of 8 tasks
chore(🤖): Bump
NVIDIA/Megatron-LM
to b8dce17...
(2025-03-26)
Run CICD
#12780
opened Mar 26, 2025 by
ko3n1g
Loading…
1 task
Fix TransformerBlock cuda_graphs compatibility with MCore
Run CICD
#12779
opened Mar 26, 2025 by
buptzyb
Loading…
ci: Measure multiprocessing
CI
no-fail-fast
Run CICD
#12778
opened Mar 25, 2025 by
ko3n1g
Loading…
8 tasks
Adds Standard transformer flops formula to flops callback
common
#12774
opened Mar 25, 2025 by
jomitchellnv
Loading…
5 tasks
Use first_value decorator for triton_infer_fn plus minor changes
skip-linting
#12773
opened Mar 25, 2025 by
janekl
Loading…
8 tasks
add finetune support for Auto Configurator
#12770
opened Mar 25, 2025 by
dimapihtar
Loading…
8 tasks
Adding more doc-strings to megatron_parallel.py
#12767
opened Mar 25, 2025 by
marcromeyn
Loading…
8 tasks
chore(🤖): Bump
NVIDIA/Megatron-LM
to cdbb175...
(2025-03-25)
Run CICD
#12766
opened Mar 25, 2025 by
ko3n1g
Loading…
1 task
[automodel][WIP]Add linear cross entropy loss
#12760
opened Mar 24, 2025 by
yuanzhedong
Loading…
8 tasks
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.