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

huggingface上openllama-13b的模型大小为26.4G,转换为huggingface那种模型格式之后模型大小为24.7G,这也就是大概是以fp16或者是bf16保存的模型 #119

Open
belle9217 opened this issue Jul 18, 2023 · 0 comments

Comments

@belle9217
Copy link

huggingface上openllama-13b的模型大小为26.4G,转换为huggingface那种模型格式之后模型大小为24.7G,这也就是大概是以fp16或者是bf16保存的模型,那这样的话,请问我在加载这个权重时候需要设置什么参数吗,还是直接加载这种半精度的权重进行训练?ps:我试了直接用半精度的权重进行训练,模型的生成结果重复的比较多,我在想是不是要把权重转换为fp32那种,还是全程都用fp16那种进行训练,因此影响了模型的精度。

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

1 participant