flowaicom/Flow-Judge-v0.1-binary-heldout
收藏Hugging Face2024-09-18 更新2025-04-08 收录
下载链接：
https://hf-mirror.com/datasets/flowaicom/Flow-Judge-v0.1-binary-heldout
下载链接
链接失效反馈官方服务：
资源简介：
这是一个用于评估Flow-Judge-v0.1模型在生成任务评估方面的能力的数据集，包含316个合成实例，这些实例用于在广泛的评估标准下测试模型正确评估生成任务的能力。
This dataset is used to evaluate the ability of the Flow-Judge-v0.1 model to correctly assess generation tasks across a broad set of criteria, containing 316 synthetic instances for testing the models performance.
提供机构：
flowaicom
