Skip to content

Issues: NVIDIA/NeMo-Aligner

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.

Sign up for GitHub

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

67 Open 16 Closed
67 Open 16 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Out of Memory (OOM) During Training a LLaMA 7B Reward Model (8 A800 40GB GPUs) bug Something isn't working
#444 opened Dec 11, 2024 by qingyiaaaaa
use lightning or pytorch-lightning bug Something isn't working
#438 opened Dec 10, 2024 by better629
5
OOM when saving torch_dist checkpoint bug Something isn't working
#436 opened Dec 7, 2024 by Cppowboy
5
NeMo Trainer / PTL Based Trainer
#426 opened Nov 30, 2024 by ashvinnihalani
Fix dev branch's build after PTL upgrade bug Something isn't working
#418 opened Nov 22, 2024 by terrykong
@ko3n1g @maanug-nv
Will you support online DPO?
#414 opened Nov 21, 2024 by Shiguang-Guo
How can I use nvidia/Llama-3.1-Nemotron-70B-Reward-HF directly for inference?
#360 opened Oct 25, 2024 by arunasank
4
serve_reward_model goes down bug Something isn't working
#351 opened Oct 18, 2024 by AtsunoriFujita
3
attribute_annotate.py is not worked by KeyError: 'exceeded' bug Something isn't working
#349 opened Oct 18, 2024 by AtsunoriFujita
Unable to pip install nemo-aligner
#342 opened Oct 11, 2024 by SCccc21
@ko3n1g
1
[Question] Converting a Megatron-LM ckpt to nemo so we can use NeMo-Aligner for post-training
#340 opened Oct 10, 2024 by abgoswam
Error during saving checkpoint with TensorRT-enabled PPO actor training bug Something isn't working
#281 opened Sep 5, 2024 by haizadinia
2
[Question] TransfomerEngine and Apex dependencies bug Something isn't working
#278 opened Sep 2, 2024 by peri044
make build_dataloader not take in cfg
#273 opened Aug 28, 2024 by gshennvm
common class for aligner models
#272 opened Aug 27, 2024 by gshennvm
Request for Context Parallel Support in MegatronGPTDPOModel
#271 opened Aug 27, 2024 by Wolfwjs
Does NeMo Aligner support tensor parallel and pipeline parallel?
#265 opened Aug 15, 2024 by cizhenshi
GPTGenerateTRTLLM.trt_llm_exporter.refit failed due to empty weights in the refit engine during PPO actor training bug Something isn't working
#264 opened Aug 10, 2024 by renweizhukov
1
job hangs or IndexError when train reward model with PP> 1 bug Something isn't working
#251 opened Jul 24, 2024 by zirui
@ashors1
6
How to shuffle data before the start of each epoch?
#250 opened Jul 24, 2024 by Cppowboy
SFT not working on nemo:24.05.01 container bug Something isn't working
#236 opened Jul 13, 2024 by vecorro
3
better add_BOS and add_EOS support in reward models
#231 opened Jul 10, 2024 by gshennvm
reward-bench for Reward Model
#230 opened Jul 6, 2024 by lss11005
1
Policy Log Probs and Reference Log Probs differ at 1st iteration of DPO/RPO bug Something isn't working
#227 opened Jul 3, 2024 by shengyangs
@trias702 @gshennvm @shengyangs
LoRA for Reward Model Training
#225 opened Jul 2, 2024 by bugsz
ProTip! Follow long discussions with comments:>50.