budecosystem/GPT4-Mixtral-Judge-Battles-100K-Complexity-train

这是一个包含对话生成的数据集，其中包含了两个模型（model_a和model_b）生成的响应（response_a和response_b），以及对应的提示（prompt）。数据集还包含了判断哪个模型获胜（winner_model_a和winner_model_b）或者是否平局（winner_tie）的信息。此外，还包含了每个示例的复杂性（complexity）评分。数据集的总大小为291,530,631字节，训练集包含109,101个示例。

This is a dataset containing dialogue generation, which includes responses generated by two models (model_a and model_b) corresponding to the prompts (prompt). The dataset also includes information on which model won (winner_model_a and winner_model_b) or if it was a tie (winner_tie). Additionally, it includes a complexity score for each example. The total size of the dataset is 291,530,631 bytes, with the training set containing 109,101 examples.