collinear-ai/support-tickets-phi-sampling-random-claude-judged

该数据集包含多个特征，如问题区域、问题类别、客户情绪、产品类别、对话内容等。此外，还包含生成项、提取的对话、正确性评分、自然性评分等序列数据。数据集被分割为多个子集，每个子集对应不同的模型或配置，并提供了每个子集的字节大小和示例数量。

The dataset contains multiple features such as issue area, issue category, customer sentiment, product category, conversation content, etc. Additionally, it includes sequences of generated items, extracted conversations, correctness scores, naturalness scores, etc. The dataset is divided into several subsets, each corresponding to different models or configurations, and provides the byte size and number of examples for each subset.