Skip to content

Issues: microsoft/DeepSpeed

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

982 Open 1,645 Closed
982 Open 1,645 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

[BUG] 1-bit LAMB not compatible with bf16 bug Something isn't working training
#5708 opened Jun 28, 2024 by catid
[BUG] Fine-tuned model outputs are empty. bug Something isn't working training
#5706 opened Jun 28, 2024 by IYIAscension
on Activation Checkpointing bug Something isn't working training
#5704 opened Jun 28, 2024 by ChaunceyWang
6
[BUG] Mixed-precision: fp16 will cast input_ids into torch.cuda.HalfTensor instead of Long or Int.
#5701 opened Jun 28, 2024 by zhaoyang02
[REQUEST]win10 install fail build_win.bat
#5698 opened Jun 26, 2024 by shark-xiake
@costin-eseanu
2
Tensor（hidden states）missing across GPU in Pipeline Parallelism Training[BUG] bug Something isn't working training
#5696 opened Jun 25, 2024 by Youngluc
@tohtana
[BUG] Regression: 0.14.3 causes grad_norm to be zero bug Something isn't working training
#5692 opened Jun 21, 2024 by rosario-purple
@samadejacobs
1
[ERROR] [launch.py:321:sigkill_handler] exits with return code = -11 bug Something isn't working training
#5690 opened Jun 21, 2024 by shag1802
@duli2012
Running out of CPU memory. Dataset is loaded for each created process bug Something isn't working training
#5689 opened Jun 21, 2024 by MikeMitsios
@umchand
[BUG] inference ValueError bug Something isn't working inference
#5685 opened Jun 19, 2024 by zxrneu
@HeyangQin
[BUG] Logs full of FutureWarning when training with nightly PyTorch bug Something isn't working training
#5682 opened Jun 18, 2024 by rosario-purple
@loadams
1
[BUG] Using and Building DeepSpeedCPUAdam bug Something isn't working training
#5677 opened Jun 18, 2024 by oabuhamdan
@loadams
23
Getting parameters of embeddings （safe_get_local_fp32_param）and setting the weight of embeddings （safe_set_local_fp32_param） does not work (bug?).
#5667 opened Jun 17, 2024 by Git-Shaw
@samadejacobs
AssertionError: Unable to pre-compile ops without torch installed. Please install torch before attempting to pre-compile ops.
#5663 opened Jun 14, 2024 by nitinmukesh
@loadams
6
[BUG] 'Invalidate trace cache' with Seq2SeqTrainer+predict_with_generate+Zero3 bug Something isn't working inference
#5662 opened Jun 14, 2024 by Osterlohe
@jomayeri
does DeepSpeed support AMSP (a new DP shard strategy) enhancement New feature or request
#5661 opened Jun 14, 2024 by guoyejun
Fail to use zero_init to construct llama2 with deepspeed zero3 and bnb!
#5660 opened Jun 14, 2024 by CHNRyan
@lekurile
RuntimeError: Error building extension 'cpu_adam', because /usr/bin/ld: can not find -lcurand，help!
#5659 opened Jun 14, 2024 by hekaijie123
2
[BUG] Running llama2-7b step3 with tensor parallel and HE fails due to incompatible shapes bug Something isn't working deepspeed-chat Related to DeepSpeed-Chat
#5656 opened Jun 13, 2024 by ShellyNR
[BUG] oneapi/ccl.hpp: No such file or directory. bug Something isn't working training
#5653 opened Jun 12, 2024 by weiji14
@loadams
1
RuntimeError: still have inflight params[BUG] bug Something isn't working training
#5648 opened Jun 12, 2024 by iszengxin
1
Inference with the MoE based GPT model trained by ds_pretrain_gpt_345M_MoE128.sh [BUG] bug Something isn't working inference
#5647 opened Jun 12, 2024 by haoranlll
[BUG] File not found in autotuner cache in multi-node setting on SLURM bug Something isn't working training
#5646 opened Jun 12, 2024 by jubueche
@loadams
1
Why doesn't deepspeed stage 3 allow a batch size of 1 with multiple GPUs? bug Something isn't working training
#5645 opened Jun 12, 2024 by AceMcAwesome77
[BUG] RuntimeError encountered when generating tokens from a Meta-Llama-3-8B-Instruct model initialized with 4-bit or 8-bit quantization bug Something isn't working compression
#5644 opened Jun 11, 2024 by Atry
2
ProTip! Mix and match filters to narrow down what you’re looking for.