Cornell-AGI/gsm8k_size_qwen2.5_7b_eval

该数据集是一个包含多个特征字段的集合，用于训练机器学习模型。特征包括数据来源、提示（内容与角色）、能力、奖励模型（真实值与风格）、额外信息（答案、索引、问题、分割）、多个响应和评估分数。数据集包含一个训练集，共有7473个示例。

This dataset is a collection of multiple feature fields designed for training machine learning models. Features include data source, prompt (content and role), ability, reward model (ground truth and style), extra information (answer, index, question, split), multiple responses, and evaluation scores. The dataset contains a training set with a total of 7473 examples.