Skip to content

Issues: NVIDIA/TransformerEngine

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

111 Open 140 Closed
111 Open 140 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

Get Stuck at Building Wheel build Build system
#976 opened Jun 27, 2024 by kingformatty
1
no boost in performance with Ada GPU
#972 opened Jun 26, 2024 by saurabh-kataria
tp_overlap need tensor parallel is equal world size ?
#966 opened Jun 25, 2024 by kuangdao
2
How to cast 16/32-bit to FP8? question Further information is requested
#965 opened Jun 25, 2024 by mxjmtxrm
3
nan loss when training in fp8 with rotary embedding
#962 opened Jun 25, 2024 by saurabh-kataria
2
Why is the result of context-parallel DotProductAttention influenced by the random seed?
#961 opened Jun 25, 2024 by LitPrice
create_communicator_grouped2 may trigger uninit value memory issue(randomly crash) when you train more iterations.
#959 opened Jun 24, 2024 by anderson101866
NaN loss issues when I switch to the Transformer Engine TransformerLayer from pytorch layer
#953 opened Jun 21, 2024 by jasonkrone
AttnFuncWithCP can use less memory
#952 opened Jun 21, 2024 by i4never
1
how to use FusedRMSNorm? question Further information is requested
#948 opened Jun 19, 2024 by EthanChen1234
1
Why use two streams for context parallel
#947 opened Jun 19, 2024 by Edenzzzz
2
backward fails after updating to main branch TE
#941 opened Jun 18, 2024 by 1049451037
[BUG] Assertion failed: t.data.dptr != nullptr. Input x is not allocated! bug Something isn't working
#935 opened Jun 17, 2024 by alexdremov
1
'TEDotProductAttention' object has no attribute 'tp_group_initialized' bug Something isn't working
#934 opened Jun 17, 2024 by 1049451037
2
How to install with CuDNN 9.0+ ?
#930 opened Jun 14, 2024 by tianyan01
@cyanguwa
2
[Feature Request][PyTorch] Support thd format for fp8 tensors in DotProductAttention
#923 opened Jun 13, 2024 by alexdremov
How to use FP8 of TransformerEngine in inference
#922 opened Jun 13, 2024 by Godlovecui
2
Compiling on Slurmcluster fatal error: cudnn.h: No such file or directory bug Something isn't working build Build system
#918 opened Jun 12, 2024 by windprak
3
The precision is not aligned by the index
#909 opened Jun 11, 2024 by Amanda-Barbara
What is padding_causal?
#908 opened Jun 11, 2024 by 1049451037
[Feature Request] Integration of DiT components into TransformerEngine.
#900 opened Jun 10, 2024 by okotaku
Can't find nvToolsExt during build
#879 opened May 29, 2024 by kvablack
1
import transformer_engine initializes CUDA bug Something isn't working
#872 opened May 28, 2024 by szmigacz
1
Strange behavior when import torch after import te. bug Something isn't working
#871 opened May 27, 2024 by GGGGGGXY
1
Port down up may cause hang when using TE in training.
#866 opened May 24, 2024 by holmes313
ProTip! Updated in the last three days: updated:>2024-06-25.