yulia-volkova/mmlu-chua-no-reasoning-summary

该数据集包含143个训练样本，每个样本包含以下特征：pi（整数类型）、question（字符串类型，表示问题）、gt_answer（字符串类型，表示正确答案）、cue_answer（字符串类型，表示提示答案）、accuracy（浮点数类型，表示准确率）、n_valid_responses（整数类型，表示有效回答数量）、n_correct（整数类型，表示正确回答数量）和most_common_given_answer（字符串类型，表示最常见的回答）。数据集未明确描述其具体用途或内容，但结构表明它可能用于问答或回答准确性分析任务。

The dataset contains 143 training examples, each with the following features: pi (integer), question (string, representing the question), gt_answer (string, representing the ground truth answer), cue_answer (string, representing the cued answer), accuracy (float, representing accuracy), n_valid_responses (integer, representing the number of valid responses), n_correct (integer, representing the number of correct responses), and most_common_given_answer (string, representing the most common given answer). The dataset does not explicitly describe its specific purpose or content, but its structure suggests it may be used for question-answering or answer accuracy analysis tasks.