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

977 Open 1,632 Closed
977 Open 1,632 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
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

How to set different learning rates for different parameters of LLMs
#5665 opened Jun 15, 2024 by jpWang
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
[BUG] 'Invalidate trace cache' with Seq2SeqTrainer+predict_with_generate+Zero3 bug Something isn't working inference
#5662 opened Jun 14, 2024 by Osterlohe
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
RuntimeError: Error building extension 'cpu_adam', because /usr/bin/ld: can not find -lcurand，help!
#5659 opened Jun 14, 2024 by hekaijie123
[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
1
RuntimeError: still have inflight params[BUG] bug Something isn't working training
#5648 opened Jun 12, 2024 by iszengxin
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
[BUG] 1 line logic issue: flipped sign/direction in _partition_param_sec of partition_parameters.py? bug Something isn't working training
#5642 opened Jun 11, 2024 by dukleryoni
1
[BUG] tortoise_tts.py fails on deepspeed/pydantic error bug Something isn't working inference
#5641 opened Jun 11, 2024 by tholonia
@loadams
Does deepspeed support aarch64?
#5640 opened Jun 11, 2024 by khayamgondal
[HELP] How to safely switch trainable parameters in ZeRO-3 stage?
#5639 opened Jun 11, 2024 by Ledzy
Deepspeed zero3 + qlora arise problem! Params didn't sharded first before load to each GPU!
#5637 opened Jun 11, 2024 by CHNRyan
[BUG] 4-bit quantized models would repeatedly generate the same tokens when bf16.enabled is true bug Something isn't working compression
#5636 opened Jun 10, 2024 by Atry
1
Deepspeed stage 3 hanging after 1st validation sample bug Something isn't working training
#5635 opened Jun 10, 2024 by AceMcAwesome77
[BUG] RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! bug Something isn't working training
#5634 opened Jun 10, 2024 by fahadh4ilyas
1
[BUG] is_zero_init_model is always False when I'm using zero_init! bug Something isn't working training
#5631 opened Jun 8, 2024 by CHNRyan
4
[BUG] RuntimeError encountered when generating tokens from a DeepSpeedHybridEngine initialized with 4-bit quantization. bug Something isn't working deepspeed-chat Related to DeepSpeed-Chat
#5630 opened Jun 8, 2024 by Atry
2
[BUG] 1: error: must run as root and 2: raise RuntimeError("Ninja is required to load C++ extensions") bug Something isn't working training
#5627 opened Jun 7, 2024 by YangBrooksHan
[BUG] RuntimeError: Error building extension 'fused_adam' Loading extension module fused_adam bug Something isn't working compression
#5623 opened Jun 6, 2024 by JinQiangWang2021
ProTip! Updated in the last three days: updated:>2024-06-12.