-
Notifications
You must be signed in to change notification settings - Fork 833
Issues: huggingface/accelerate
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
save_state removes shared weights but load_state cannot load properly
#2807
opened May 27, 2024 by
MiladInk
2 of 4 tasks
Error with Deepspeed and dataloader_drop_last=True when batch size doesn't divide evenly
#2801
opened May 25, 2024 by
mrbesher
2 of 4 tasks
RuntimeError: Expected is_sm80 || is_sm90 to be true, but got false.
#2799
opened May 23, 2024 by
mostafamdy
4 tasks
Saving deepspeed ZERO-3 finetuned model fails sometimes.
#2797
opened May 23, 2024 by
xuanyaoming
2 of 4 tasks
replacing torch.utils.checkpoint with deepspeed.runtime.activation_checkpointing.checkpointing does not work
#2792
opened May 20, 2024 by
vkaul11
2 of 4 tasks
GPU Memory Imbalance and OOM Errors During Training
#2789
opened May 17, 2024 by
DONGRYEOLLEE1
2 of 4 tasks
[DeepSpeed] Asking for feedback when training with zero2 with accelerate and diffusers
#2787
opened May 16, 2024 by
sayakpaul
AcceleratorState
object has no attribute distributed_type
.
#2786
opened May 16, 2024 by
evelinamorim
2 of 4 tasks
Unable to launch DeepSpeed multinode training with a heterogenous mix of # devices per node.
#2780
opened May 14, 2024 by
iantbutler01
2 of 4 tasks
Unable to load mistralai/Mixtral-8x7B-Instruct-v0.1 using mps
#2778
opened May 14, 2024 by
chimezie
2 of 4 tasks
Accelerate FSDP RuntimeError: Tensors of the same index must be on the same device and the same dtype
#2764
opened May 10, 2024 by
yaswanthchittepu
Cuda Out of memory while loading PEFT weights using accelerate on multi gpu
#2760
opened May 10, 2024 by
sidtandon2014
2 of 4 tasks
Performance on single GPU is much better than on Multi-GPUs
#2754
opened May 8, 2024 by
baicenxiao
3 of 4 tasks
PicklingError: Can't pickle <function Embedding.forward at XXXXXXX> it's not the same object as torch.nn.modules.sparse.Embedding.forward
#2749
opened May 7, 2024 by
arpit2665
1 of 4 tasks
4-bit quantization cannot load weights to meta device for bias terms of the linear layer: NotImplementedError: Cannot copy out of meta tensor; no data!
#2742
opened May 5, 2024 by
MuhammedHasan
2 of 4 tasks
2
Previous Next
ProTip!
Exclude everything labeled
bug
with -label:bug.