Ayush-Singh/reward-bench-Phi-3-mini-128k-instruct-set2-scores
收藏Hugging Face2025-02-13 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/Ayush-Singh/reward-bench-Phi-3-mini-128k-instruct-set2-scores
下载链接
链接失效反馈官方服务：
资源简介：
这是一个包含多种split的数据集，用于评估模型在不同场景下的性能。每个split包含多个字段，如prompt、chosen、chosen_model等，用于记录模型的输入、输出、选择的模型、得分和解释等。数据集包含alpacaeval_hard、hep_go等多种split，每个split有不同的示例数量和大小。
This is a dataset with multiple splits designed for evaluating model performance in various scenarios. Each split contains fields such as prompt, chosen, chosen_model, etc., to record the models input, output, chosen model, scores, and justifications. The dataset includes splits like alpacaeval_hard, hep_go, each with a different number of examples and size.
提供机构：
Ayush-Singh
