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

pytorch model & ring attention #55

Open
LzhinFdu opened this issue Mar 5, 2024 · 4 comments
Open

pytorch model & ring attention #55

LzhinFdu opened this issue Mar 5, 2024 · 4 comments

Comments

@LzhinFdu
Copy link

LzhinFdu commented Mar 5, 2024

Thanks for sharing this excellent great work. We want to use pytorch models to try the effect of ring attention. Are there any plans to develop ring attention implementation under pytorch?

@apexspyche
Copy link

What do you have in mind? Is this model suitable for tokenized ecosystem and bridging liquidity and creating a smart algorithm for bridging / blending / mending and growth hacking liquidity across and between multiple TOKENS

@kabachuha
Copy link

Lucidrains has a pytorch implementation of RingAttention https://github.com/lucidrains/ring-attention-pytorch

@LzhinFdu
Copy link
Author

LzhinFdu commented Mar 5, 2024

Lucidrains has a pytorch implementation of RingAttention https://github.com/lucidrains/ring-attention-pytorch

Have you tried this repo? I don’t know whether the experimental results are as expected.
Seems that the model posted on huggingface cannot use it directly to call ring attention.

@LzhinFdu
Copy link
Author

LzhinFdu commented Mar 5, 2024

What do you have in mind? Is this model suitable for tokenized ecosystem and bridging liquidity and creating a smart algorithm for bridging / blending / mending and growth hacking liquidity across and between multiple TOKENS

I just want to call ring attention when using the trained pytorch LLM for inference.

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

No branches or pull requests

3 participants