-
Notifications
You must be signed in to change notification settings - Fork 2.3k
Issues: NVIDIA/Megatron-LM
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
[QUESTION] Adding a new parameter in ColumnParallelLinear/RowParallelLinear raises Error
#1150
opened Sep 19, 2024 by
haolibai
[BUG] Learning rate not overrided when set
--override-opt_param-scheduler
#1138
opened Sep 13, 2024 by
TissueC
[ENHANCEMENT] Preprocessing data that is already partitioned and gzipped
#1135
opened Sep 13, 2024 by
ianporada
[BUG] 'NoneType' object has no attribute 'shape' error raised when saving model state with the pretrain_gpt.py
#1134
opened Sep 13, 2024 by
hwang2006
[QUESTION] Possible to install "from userlib.auto_resume import AutoResume"
#1133
opened Sep 11, 2024 by
orrzohar
[BUG]"Unexpected key(s) in state_dict" while loading Llama-megatron checkpoint.
#1132
opened Sep 11, 2024 by
mxjmtxrm
Dose Context Parallel support Packing Inputs Without Cross-Contamination Attention?
#1131
opened Sep 11, 2024 by
Lzhang-hub
[QUESTION] How to Identify Which Document Processing in Forward
#1128
opened Sep 9, 2024 by
zixianwang2022
[QUESTION] Epochs Larger Than 1 When Specified with Trained Samples
#1127
opened Sep 9, 2024 by
zixianwang2022
[QUESTION] tensor_parallel.broadcast_data and train_valid_test_datasets_provider.is_distributed = True
#1125
opened Sep 9, 2024 by
KookHoiKim
[BUG]
run_simple_mcore_train_loop.py
bugs when moditied tensor_model_parallel_size
from 2
to 1
#1038
opened Aug 27, 2024 by
1195343015
[BUG] When the model has extra layers, initializing the model from dist-ckpt results in an error
#1032
opened Aug 26, 2024 by
haolin-nju
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.