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.

Sign up for GitHub

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

Jump to bottom

我们如何针对扩展区块微调? #3

Open
win10ogod opened this issue Jan 10, 2024 · 5 comments
Open

我们如何针对扩展区块微调? #3

win10ogod opened this issue Jan 10, 2024 · 5 comments

Comments

@win10ogod
Copy link

我们如何针对扩展区块微调?
@win10ogod
Copy link
Author

@hills-code How do we train for extended blocks?

@hills-code
Copy link
Collaborator

感谢关注！ 我把训练代码上传了在这个repo下面了，也可以查看https://github.com/hills-code/open-instruct/tree/llama-pro

@JohnZhuYX
Copy link

@hills-code 你好，请问我在https://github.com/hills-code/open-instruct/blob/llama-pro/scripts/finetune_codealpaca.sh脚本中没有看到你对原来的层的冻结操作，你是把所有的层都进行训练了吗？还有，你这个项目里是哪个阶段的训练？PT ? SFT?

@hills-code
Copy link
Collaborator

这个项目是SFT的训练，在这个阶段是所有参数一起训练，和普通的SFT是一致的；在Pretrain的时候会冻结参数，具体的操作在这里https://github.com/hills-code/open-instruct/blob/7c2b14d3d319028c68657946ca2c16b248f866e8/open_instruct/customized_trainer.py#L53

@JohnZhuYX
Copy link

我看了customized_trainer.py这个文件，但是这个文件只是用于扩展模型，也就是把一个7B的模型变成一个8B的模型，但是增加的块都是初始状态，没有被训练。有没有针对这些新增块（也就是冻结所有原始块）进行PT轮训练的DEMO

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@JohnZhuYX @hills-code @win10ogod