pittawat/medqa_20_long_results_v2

该数据集包含以下字段：id（字符串类型）、question（字符串类型）、ground_truth（字符串类型）、prompt（字符串类型）、response（字符串类型）、prompt_tokens（整型）、response_tokens（整型）、processing_time（浮点型）、created_at（时间戳类型）和metadata（包含dataset、model和prompt_template的字典类型）。数据集分为训练集，共有24个样本。该数据集用于训练和评估模型对问题的回答能力。

The dataset includes the following fields: id (string), question (string), ground_truth (string), prompt (string), response (string), prompt_tokens (integer), response_tokens (integer), processing_time (float), created_at (timestamp), and metadata (a dictionary containing dataset, model, and prompt_template). The dataset is split into a training set with a total of 24 samples. It is used for training and evaluating the models ability to answer questions.