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

微调的一些问题 #107

Open
zhouzhou0322 opened this issue Nov 9, 2023 · 3 comments
Open

微调的一些问题 #107

zhouzhou0322 opened this issue Nov 9, 2023 · 3 comments
Assignees
Labels
bug Something isn't working

Comments

@zhouzhou0322
Copy link

🐛 bug 说明

  1. 请问微调m3e-base需要多少数据量才有效果,我使用条左右训练,貌似没有效果,微调前后embedding的l2距离是一样
  2. 微调时候loss为0
  3. 合同签订前 ,合同签订后。 这两个目前我使用m3e 他们的语义很近似,但是在我的业务场景他们应该是最不接近的。微调的很多数据也是类似的后缀的意思不一样 但是总体的语义是接近的。这样可以微调吗?

谢谢

Python Version

None

@zhouzhou0322 zhouzhou0322 added the bug Something isn't working label Nov 9, 2023
@wangyuxinwhy
Copy link
Owner

  1. 一般至少数百条数据
  2. 微调主要应该看验证集的 Loss 不要看训练集的 Loss
  3. 可以微调的

@suna0922
Copy link

  1. 一般至少数百条数据
  2. 微调主要应该看验证集的 Loss 不要看训练集的 Loss
  3. 可以微调的

您好,请问在微调时,不同数据量的数据选择微调策略(全参数/bitfit)有什么建议嘛~

@wangyuxinwhy
Copy link
Owner

我建议最好都是全参数微调,bitfit 是为了微调 7B 以上的模型,但这个算法的效果我自己实验下来不是很理想...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

3 participants