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

替换LLM重新进行训练 #796

Open
SuHuaFeng opened this issue Dec 25, 2024 · 0 comments
Open

替换LLM重新进行训练 #796

SuHuaFeng opened this issue Dec 25, 2024 · 0 comments

Comments

@SuHuaFeng
Copy link

我现在把llm换成了qwen2-coder-7b,这是一个代码生成模型,然后我进行了第一阶段的训练,我使用20000步的checkpoint推理发现结果很差(如下图),
image
我觉得可能是训练的步数不够也可能是我的训练数据格式就是错的,下图是我目前的训练数据格式,输入是图像和指令,gpt输出是代码,
image
看了论文,第一阶段做的是对比学习,输入是不是应该是图像和代码,然后gpt的输出应该是一个标签,0或者1,代表图像和代码是否匹配,然后之后进行第二阶段的全参微调才使用我目前这样的格式。
希望您能解答一下我的疑问,应该怎么修改数据格式,然后我的替换llm的这个想法在实现的过程需要注意些什么,万分感激!

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