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

llama3 8B中文微调模型Llama3-Chinese-8B-Instruct 存在对生成的答案一直重复的问题 #330

Open
Ryan-0805 opened this issue Apr 28, 2024 · 9 comments

Comments

@Ryan-0805
Copy link

提问任何问题,都会一直重复自己的答案,直到达到模型的max_token;

@TimchaStudio
Copy link

遇到同样的bug

image

@Zhouxinyu668
Copy link

我也遇到了同样的问题
image

@Zhouxinyu668
Copy link

作者不考虑回应一下吗?

@hito0512
Copy link

hito0512 commented May 8, 2024

image
@TimchaStudio 这个是用哪个框架部署的?

@TimchaStudio
Copy link

@hito0512 chat.oaiui.com

这个问题还没有解决

@ZhichengQian1
Copy link

我尝试了一下,也存在这个问题

@nbet
Copy link

nbet commented Jun 3, 2024

me too too~~

@3h2b
Copy link

3h2b commented Jul 9, 2024

大语言生成回答时一般会先生成一个开始符号类似于《start》然后输出正文,紧接着在结束时输出一个类似于《end》的特殊字符作为判断此句话结束。在这个具体问题里结束符用的可能是《\s》。出现这个问题可能是模型在训练中文语料库时更换了结束字符导致原来判断生成结束的符号失去作用。大伙可以自行添加判断生成结束的标识符可以解决这个问题。

@chhaig
Copy link

chhaig commented Aug 18, 2024

调整 temperature的大小,设置小于0.3以下。

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

8 participants