Skip to content
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

[Bug] Fix a bug in fused rope #1750

Merged
merged 1 commit into from
Nov 16, 2023
Merged

[Bug] Fix a bug in fused rope #1750

merged 1 commit into from
Nov 16, 2023

Conversation

yaox12
Copy link
Contributor

@yaox12 yaox12 commented Nov 15, 2023

Fix a bug in the fused rope backward calculation. cc @crcrpar

@crcrpar crcrpar added this to the 23.12 milestone Nov 16, 2023
@crcrpar crcrpar merged commit fd4ae7d into NVIDIA:master Nov 16, 2023
@yaox12 yaox12 deleted the fix_fused_rope branch November 17, 2023 01:10
StellaAthena added a commit to EleutherAI/gpt-neox that referenced this pull request Nov 25, 2023
Ports the fix from NVIDIA/apex#1750 into this branch.
Quentin-Anthony added a commit to EleutherAI/gpt-neox that referenced this pull request Jan 5, 2024
* Create fused_rotary_positional_embedding.cpp

* Create fused_rotary_positional_embedding.h

* Create fused_rotary_positional_embedding_cuda.cu

* Update fused_rotary_positional_embedding.h

Ports the fix from NVIDIA/apex#1750 into this branch.

* Update neox_args.py

* Update setup.py

* Update initialize.py

* Update setup.py

* Update __init__.py

* Update test_fused_kernels.py

* Update setup.py

* Create fused_rope.py

* Update fused_rotary_positional_embedding.h

* Update fused_rotary_positional_embedding.cpp

* Update fused_rotary_positional_embedding.cpp

* Update transformer.py

* Update transformer.py

Just checked and this should work for bf16. Or, at least, the reason I originally thought it wouldn't doesn't apply.

* Update transformer.py

* Create 125M_fused_rope.yml

* Update 125M_fused_rope.yml

* Update transformer.py

Add `self.rope_fusion = neox_args.rope_fusion` so that `ParallelSelfAttention` knows if we're using rope fusion.

* Update NeoXArgs docs automatically

* Update NeoXArgs docs automatically

* Fix fused rope

Just needed to bring in the latest headers/sources,
and call into it the right way from transformers.py.

* Add rope_fusion arg to all ymls

---------

Co-authored-by: Stella Biderman <[email protected]>
Co-authored-by: github-actions <[email protected]>
Co-authored-by: Quentin Anthony <[email protected]>
Co-authored-by: Yang Zhang <[email protected]>
caaatch22 pushed a commit to caaatch22/apex that referenced this pull request Dec 18, 2024
SkyHeroesS pushed a commit to SkyHeroesS/beforeholiday that referenced this pull request Dec 20, 2024
caaatch22 pushed a commit to caaatch22/apex that referenced this pull request Dec 23, 2024
caaatch22 pushed a commit to caaatch22/apex that referenced this pull request Dec 23, 2024
caaatch22 pushed a commit to caaatch22/apex that referenced this pull request Dec 23, 2024
caaatch22 pushed a commit to caaatch22/apex that referenced this pull request Dec 23, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants