nate-rahn/sexism_filter_prompt_claude_3_7_sonnet_concurrent_10x_batch_judgment-judgment-input

该数据集包含成对的查询（query1和query2），以及它们对应的类别（category1和category2）、级别（level1和level2）、配对类型（pair_type）、原则（principle）、完整提示（full_prompt）和预期偏好（expected_preference）。此外，还包括时间戳（timestamp）和自定义ID（custom_id）。数据集主要用于训练模型以比较或评估查询之间的偏好。The dataset contains pairs of queries (query1 and query2), along with their corresponding categories (category1 and category2), levels (level1 and level2), pair types (pair_type), principles (principle), full prompts (full_prompt), and expected preferences (expected_preference). Additionally, it includes a timestamp (timestamp) and a custom ID (custom_id). The dataset is primarily used for training models to compare or evaluate preferences between queries.