mlfoundations-dev/nemotron-sft_300000_1742858011_eval_0981
收藏Hugging Face2025-03-25 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/mlfoundations-dev/nemotron-sft_300000_1742858011_eval_0981
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是mlfoundations-dev/nemotron-sft_300000_1742858011_eval_0981,包含了针对多个任务(AIME24,AIME25,AMC23,MATH500,GPQADiamond,LiveCodeBench)的预计算模型输出,用于评估模型在这些任务上的表现。每个任务都有多次运行的结果,结果包括准确率、解决的问题数以及总问题数。
This dataset is mlfoundations-dev/nemotron-sft_300000_1742858011_eval_0981, which includes precomputed model outputs for multiple tasks (AIME24, AIME25, AMC23, MATH500, GPQADiamond, LiveCodeBench) used for evaluating the performance of the model on these tasks. Each task has multiple run results, including accuracy, number of questions solved, and total number of questions.
提供机构:
mlfoundations-dev



