budecosystem/Llama3.3Gemma2-WALPAM-Preference-30K-Complexity-train
收藏Hugging Face2025-01-17 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/budecosystem/Llama3.3Gemma2-WALPAM-Preference-30K-Complexity-train
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含模型A和模型B在不同领域的提示（prompt）下的性能比较结果。每个样本包括两个模型的标识符（model_a和model_b）、领域（domain）、提示文本（prompt）、模型A和模型B的胜率（winner_model_a和winner_model_b）、是否平局（tie_ab）以及问题的复杂性（complexity）。数据集分为训练集，共有29928个示例，文件大小为5729848字节。
The dataset contains the performance comparison results of Model A and Model B under different domain prompts. Each sample includes identifiers for the two models (model_a and model_b), the domain, the prompt text, the win rates of Model A and Model B (winner_model_a and winner_model_b), whether there is a tie (tie_ab), and the complexity of the issue. The dataset is split into a training set with a total of 29,928 examples, with a file size of 5,729,848 bytes.
提供机构：
budecosystem
