mosama/final_grpo_data

该数据集包含107,067个训练示例，用于AI模型训练，特别关注任务执行、工具使用和答案生成。每个示例包括提示、RL任务、预期动作、工具列表、内部LaTeX答案、预期答案、选项、模式字符串、验证状态和问题等特征字段。数据集大小为710 MB，下载大小为196 MB，适用于自然语言处理和强化学习应用。

This dataset contains 107,067 training examples for AI model training, with a focus on task execution, tool usage, and answer generation. Each example includes features such as prompt, rl_task, expected_action, tools, answer_inner_latex, expected_answer, options, schema_str, passed_validation, and question. The dataset size is 710 MB, with a download size of 196 MB, and is suitable for natural language processing and reinforcement learning applications.