lvogel123/gpqa-diamond-gpt-oss-120b

该数据集包含三个配置：results、samples和stats。results配置包含训练过程中的日志路径、评估ID、运行ID、创建时间、任务信息、任务ID、模型名称、总样本数、完成样本数、准确率、标准误和标准差等特征。samples配置包含样本ID、训练的epoch、目标、消息、多项选择题评分器的值和答案等特征。stats配置包含开始时间和完成时间、模型使用情况、输入token数、输出token数和总token数等特征。数据集分为训练集，其中results、samples和stats配置的训练集分别包含3、328和1个样本。

The dataset consists of three configurations: results, samples, and stats. The results configuration includes features such as log path, evaluation ID, run ID, creation time, task information, task ID, model name, total number of samples, number of completed samples, accuracy, standard error, and standard deviation from the training process. The samples configuration includes features like sample ID, training epoch, target, messages, multiple-choice question scorer value, and answer. The stats configuration includes features such as start and completion time, model usage, number of input tokens, number of output tokens, and total number of tokens. The dataset is split into a training set, with the results, samples, and stats configurations containing 3, 328, and 1 samples respectively.