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

Cut Your Losses in Large-Vocabulary Language Models #4794

Open
ChengzhiHuang opened this issue Nov 23, 2024 · 0 comments
Open

Cut Your Losses in Large-Vocabulary Language Models #4794

ChengzhiHuang opened this issue Nov 23, 2024 · 0 comments
Assignees

Comments

@ChengzhiHuang
Copy link
Collaborator

推荐收录

链接

https://github.com/apple/ml-cross-entropy

理由

该项目是基于论文《Cut Your Losses in Large-Vocabulary Language Models》,提出了一种新方法Cut Cross-Entropy (CCE),用于优化大词汇量语言模型的交叉熵损失计算。该方法通过只计算正确标记的logit,从而大幅减少内存需求,测试表明在小型模型中,内存消耗从24GB降至1MB,提高训练速度。

推荐人信息

如果收录这个资源,我们会在点评后面说明推荐信息人。

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

2 participants