We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
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
个人在 val 上采用 5-shot prompting 得到的基本分数异常的低(比 turbo 还糟糕)
gpt-4-1106-preview computer_network 15.78947 gpt-4-1106-preview operating_system 10.52632 gpt-4-1106-preview computer_architecture 0.00000 gpt-4-1106-preview college_programming 32.43243 gpt-4-1106-preview college_physics 36.84211 gpt-4-1106-preview college_chemistry 0.00000
@HYZ17 你们有内部试过 gpt-4-turbo 的 test 表现吗?至少我这里用 val 跑 3.5-turbo 结果和 test 蛮相近的
The text was updated successfully, but these errors were encountered:
我们进行了小部分的科目的zero-shot测试,发现gpt-4-turbo的输出格式很多样,且有时候会拒绝给出答案。这可能是准确率低的原因之一
Sorry, something went wrong.
test没有答案怎么办啊
在这个网站https://cevalbenchmark.com/ 提交获得分数
No branches or pull requests
个人在 val 上采用 5-shot prompting 得到的基本分数异常的低(比 turbo 还糟糕)
@HYZ17 你们有内部试过 gpt-4-turbo 的 test 表现吗?至少我这里用 val 跑 3.5-turbo 结果和 test 蛮相近的
The text was updated successfully, but these errors were encountered: