cemig-ceia/energy-eval-filtered

该数据集包含447个训练样本，设计用于问答或多选任务。每个样本包含以下特征：id（唯一标识符）、question_number（问题编号）、right_context（右上下文）、question（问题文本）、choices（选项列表，包含label和text）、answerKey（正确答案键）和question_with_choices（包含选项的问题文本）。数据集的结构表明它适用于需要理解上下文并回答问题的任务。

The dataset contains 447 training samples designed for question-answering or multiple-choice tasks. Each sample includes the following features: id (unique identifier), question_number (question number), right_context (right context), question (question text), choices (list of options, including label and text), answerKey (correct answer key), and question_with_choices (question text with options). The structure of the dataset suggests it is suitable for tasks that require understanding context and answering questions.