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

能不能说明一下显卡要求啊? #123

Open
pdwfree opened this issue Mar 27, 2024 · 3 comments
Open

能不能说明一下显卡要求啊? #123

pdwfree opened this issue Mar 27, 2024 · 3 comments
Assignees
Labels
enhancement New feature or request

Comments

@pdwfree
Copy link

pdwfree commented Mar 27, 2024

🚀 The feature

在说明里能不能增加一下显卡的要求啊?
比如说,哪种数据量级的数据微调时,m3e-small base large 对显卡显存的要求是什么?
4080 16G、3090 24G这些卡 单卡能跑吗?
穷人手里没有48G 80G这样的卡。
非常感谢大佬们的答复。

@pdwfree pdwfree added the enhancement New feature or request label Mar 27, 2024
@wangyuxinwhy
Copy link
Owner

16G 这种级别的卡就都够用,需要注意的是,batch_size 不要设置的太大

@susht3
Copy link

susht3 commented Apr 22, 2024

16G 这种级别的卡就都够用,需要注意的是,batch_size 不要设置的太大

您好,我的单卡是32G,但是最大只能跑batch size32;设置8卡来跑,也跑不通batch size 128,还有什么地方需要配置么?

srun -p src-12xv100-32g --workspace-id src -f pt -r N1lS.Ib.I20.8 -N 8 -d AllReduce bash finetune.sh

@wangyuxinwhy
Copy link
Owner

uniem 的显存瓶颈主要在激活上,并且依赖于 In Batch 的负采样,所以 DDP 或者 ZeRO 的方式也没有办法提升 Batch Size...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants