deepspeedai/DeepSpeedExamples

View on GitHub

Example models using DeepSpeed

Stars: 6.8k
Forks: 1.1k
Open beginner issues: 0
Indexed issues: 300
Avg merge: 2d 16h
Merged PRs (30d): 1

Dominant language: Python
License: Apache-2.0
Last GitHub push: Aug 26, 2026
Latest indexed: Sep 20, 2026
Contributing guide: No contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

300 open issues indexed

how to understand the code for calculating rewards Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#703 · 1 comment · 1 reaction · Opened Aug 22, 2023
[BUG] Evaluation sampler is assigned incorrectly in `step_2_reward_model_finetuning` Open

Difficulty 2/5 1-3 hours Newbie friendliness 48/100

deepspeedai/DeepSpeedExamples#701 · 3 reactions · Opened Aug 21, 2023
why can't I just set the 'data_sampling=true"? Open

Difficulty 3/5 1-2 days Newbie friendliness 30/100

deepspeedai/DeepSpeedExamples#696 · Opened Aug 17, 2023
Questing about why we need to add a space before prompt in stage3? Open

Difficulty 2/5 1-3 hours Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#694 · Opened Aug 17, 2023
Step 2 reward model finetuning: how is the loss computed? Open

Difficulty 4/5 3-5 days Newbie friendliness 32/100

deepspeedai/DeepSpeedExamples#693 · 1 comment · 2 reactions · Opened Aug 17, 2023
Sample codes don't work in WSL2 after #651 Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#692 · 1 comment · Opened Aug 17, 2023
Errors in cifar training and compression Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#690 · 3 comments · Opened Aug 16, 2023
How to use Tensor Parallelism for WizardCoder Inference? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#674 · Opened Aug 7, 2023
TypeError: DeepSpeedZeroOptimizer.check_overflow() got an unexpected keyword argument 'external' Open

Difficulty 3/5 1-2 days Newbie friendliness 28/100

deepspeedai/DeepSpeedExamples#673 · 4 reactions · Opened Aug 7, 2023
Deepspeed-chat feature request: specify which epoch/iteration to start EMA Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

deepspeedai/DeepSpeedExamples#670 · Opened Aug 4, 2023
[Question] Step3, fatal error: cuda_bf16.h: No such file or directory Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#669 · Opened Aug 4, 2023
LoRA problem：out of memory when 3b model with lora in 32G GPU with batchsize 2 Open

Difficulty 4/5 3-5 days Newbie friendliness 28/100

deepspeedai/DeepSpeedExamples#665 · 1 comment · Opened Aug 3, 2023
run_6.7b.sh On A6000 48G OOM Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#661 · Opened Aug 1, 2023
[Bug] Step1: Not mask source part Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#660 · 1 comment · Opened Aug 1, 2023
How to save the intermediate model? Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#654 · 1 comment · Opened Jul 21, 2023
ds_eval_config v.s. ds_config Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#653 · 2 comments · Opened Jul 19, 2023
【Update request】Deepspeed-chat's codes does not support the Hybrid Engine of DeepSpeed the version 0.9.5 Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

deepspeedai/DeepSpeedExamples#650 · 2 comments · Opened Jul 18, 2023
Tokenization extremely inefficient -- 1 example at a time appended to python list Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#649 · Opened Jul 17, 2023
inference/huggingface/text-generation example dies with: ImportError: cannot import name 'LlamaTokenizerFast' from 'transformers' Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

deepspeedai/DeepSpeedExamples#647 · Opened Jul 17, 2023
[DeepSpeed-Chat] Configuring parallelism Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

deepspeedai/DeepSpeedExamples#646 · Opened Jul 17, 2023
RuntimeError: The size of tensor a (768) must match the size of tensor b (3072) at non-singleton dimension 0 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#643 · 1 comment · Opened Jul 16, 2023
Step 3 get stuck in forward Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#642 · 1 comment · Opened Jul 14, 2023
Modifying step 1 supervised fine-tuning for non-chat datasets? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#641 · Opened Jul 14, 2023
About Chinese data training Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#640 · 1 comment · Opened Jul 14, 2023
ModuleNotFoundError: No module named 'utils.data' Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#639 · 3 comments · Opened Jul 14, 2023
Step3 Is padding side right or not? Open

Difficulty 3/5 1-2 days Newbie friendliness 32/100

deepspeedai/DeepSpeedExamples#637 · 1 comment · Opened Jul 13, 2023
[ERROR] DeepSpeed-Chat use ZeRO MiCS Configurations Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#632 · Opened Jul 10, 2023
Advice on the experiment Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#631 · Opened Jul 10, 2023
much more memory used than In theory Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#630 · 4 comments · Opened Jul 10, 2023
[Bug] In step3, a runtime error will be thrown when inference_tp_size>1 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#628 · 6 comments · Opened Jul 10, 2023
Some weights of GPTNeoXForCausalLM were not initialized from the model checkpoint Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#627 · 5 comments · Opened Jul 10, 2023
[ERROR] [launch.py:324:sigkill_handler]:Deepspeed trainng CLM issue Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#626 · Opened Jul 10, 2023
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#625 · 5 comments · 1 reaction · Opened Jul 9, 2023
"RuntimeError: The size of tensor a (5120) must match the size of tensor b (20480) at non-singleton dimension 0" in step3 Open

deepspeedai/DeepSpeedExamples#622 · 5 comments · 1 assignee · Opened Jul 3, 2023
Megatron-DeepSpeed Assertion error: last epoch number of samples exceeded max value. Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#621 · 2 comments · Opened Jul 2, 2023
LoRA in Benchmark Settings Open

deepspeedai/DeepSpeedExamples#619 · 1 comment · 1 assignee · Opened Jul 1, 2023
“Current loss scale already at minimum - cannot decrease scale anymore” when I enabled ZERO++ Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#617 · Opened Jun 30, 2023
[inference] when will deepspeed support inference with int8 gemm? Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#611 · Opened Jun 27, 2023
Training with QLORA Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#606 · 2 comments · 2 reactions · Opened Jun 25, 2023
Environment error: Error building extension 'fused_adam' Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#603 · 1 comment · Opened Jun 21, 2023
ValueError: Found `optimizer` configured in the DeepSpeed config, but no `scheduler`. Please configure a scheduler in the DeepSpeed config. Open

Difficulty 3/5 1-2 days Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#600 · Opened Jun 19, 2023
RuntimeError with BLOOMZ: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#598 · 1 comment · 2 reactions · Opened Jun 17, 2023
Do we have any plans on supporting pipeline parallel? Open

question

deepspeedai/DeepSpeedExamples#596 · 2 comments · 1 reaction · 1 assignee · Opened Jun 15, 2023
In step 3, I met a error when executing self.actor_model.eval() Open

deespeed chat hybrid engine

deepspeedai/DeepSpeedExamples#593 · 6 comments · 2 assignees · Opened Jun 13, 2023
what(): CUDA error: an illegal memory access was encountered Open

deespeed chat

deepspeedai/DeepSpeedExamples#592 · 8 comments · 1 reaction · 1 assignee · Opened Jun 13, 2023
[BUG] deepspeed-chat bloom training error, raise RuntimeError "still have inflight params " after 14 steps training of step3 with offload option turned on Open

deespeed chat new-config

deepspeedai/DeepSpeedExamples#591 · 1 assignee · Opened Jun 12, 2023
Is it right that per_device_train_batch_size = per_device_mini_train_batch_size * gradient_accumulation_steps? Open

Difficulty 5/5 Over a week Newbie friendliness 10/100

deepspeedai/DeepSpeedExamples#589 · 1 reaction · Opened Jun 12, 2023
During the training of Step 3, the reward score of my language model collapsed to a stable point Open

deespeed chat modeling

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#586 · 7 comments · Opened Jun 9, 2023
Evaluation Loader for DeepSpeed Chat Example step 2 (reward model training) Open

deespeed chat modeling

Difficulty 1/5 Under an hour Newbie friendliness 45/100

deepspeedai/DeepSpeedExamples#581 · 1 reaction · Opened Jun 8, 2023
enable critic_gradient_checkpointing, get error Open

deespeed chat question

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#578 · 3 comments · Opened Jun 7, 2023

deepspeedai/DeepSpeedExamples

Get new issues in your inbox