prometheus-eval/outcome_meta_evaluation_heuristic
收藏Hugging Face2025-01-18 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/prometheus-eval/outcome_meta_evaluation_heuristic
下载链接
链接失效反馈官方服务：
资源简介：
这是一个包含问题生成、解决和评估步骤的数据集，具体包括问题ID、问题生成器、问题描述、解决步骤、答案正确性、人工评分、模型输出和最终答案等字段。数据集分为biggen_bench、critic_bench、judge_bench和reward_bench四个部分，每个部分包含了不同类型的问题和对应的解决步骤。
This is a dataset containing problem generation, resolution, and evaluation steps, including fields such as problem ID, problem generator, problem description, resolution steps, answer correctness, human scoring, model output, and final answer. The dataset is divided into four parts: biggen_bench, critic_bench, judge_bench, and reward_bench, each containing different types of problems and corresponding resolution steps.
提供机构：
prometheus-eval
