mlfoundations-dev/Qwen2.5-7B_OpenThoughts3_eval_8179

这是一个评估数据集，包含了预计算的模型输出结果。数据集涵盖了多个数学和编程竞赛，如AIME24、AMC23、MATH500等，以及编程相关的CodeElo和CodeForces。每个数据集都有多个运行结果，包括准确率、解决的问题数和总问题数。This is an evaluation dataset containing precomputed model outputs. The dataset covers various mathematical and programming contests such as AIME24, AMC23, MATH500, and programming-related CodeElo and CodeForces. Each dataset has multiple run results including accuracy, number of problems solved, and total number of questions.