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

Internvl2.5 38B训练后如何merge checkpoints和原来的权重 #2900

Open
han3889847 opened this issue Jan 10, 2025 · 2 comments
Open

Internvl2.5 38B训练后如何merge checkpoints和原来的权重 #2900

han3889847 opened this issue Jan 10, 2025 · 2 comments

Comments

@han3889847
Copy link

您好,我用自己数据微调Internvl2.5-38B,损失下降都是正常的
输出了
image
我merge-lora使用指令
swift merge-lora
--ckpt_dir /llmcapagroup1/jtcvdata/hjn/hjn/ms-swift-main/output/InternVL2_5-38B/checkpoint-300
总是报错出现问题:
image
而且图里我没看到有intern2.5的字样

我该如何使用生成的权重合并原有38B进行LMdelpoy推理

@Jintao-Huang
Copy link
Collaborator

你看看训练和merge的swift版本是不是不一致

一个 swift3,一个swift2

@han3889847
Copy link
Author

这两个都是一套代码和环境,都是modelscope下载docker镜像,且
image
中的version.py中写着的版本是3.0.0
image

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