Chatglm3 的ceval评估结果很差！ #582
Closed
githubjqh
started this conversation in
Show and tell
Replies: 1 comment
-
|
想问一下最后是怎么解决的呢？ |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
用ceval数据集和代码测试从modelscope下载的chatglm3-6b-base模型，最后精度结果平均值为30.65%。与https://cevalbenchmark.com/static/model.html?method=ChatGLM3-6B-base中给出的结果相差太远。谁知道什么原因吗？
Beta Was this translation helpful? Give feedback.
All reactions