hsiung/llm-similarity-risk

--- dataset_info: features: - name: prompt dtype: string - name: messages list: - name: content dtype: string - name: role dtype: string splits: - name: train num_bytes: 355835764 num_examples: 59766 - name: test num_bytes: 39616427 num_examples: 8304 - name: train_random_5k num_bytes: 352570387 num_examples: 57000 - name: train_random_1k num_bytes: 347865006 num_examples: 53000 - name: train_pure_bad_high_sim_5k num_bytes: 352282888 num_examples: 57000 - name: train_pure_bad_low_sim_5k num_bytes: 352908141 num_examples: 57000 - name: train_pure_bad_high_sim_1k num_bytes: 347727585 num_examples: 53000 - name: train_pure_bad_low_sim_1k num_bytes: 348070253 num_examples: 53000 - name: train_list_high_sim_5k num_bytes: 352588208 num_examples: 57000 - name: train_list_low_sim_5k num_bytes: 352538289 num_examples: 57000 - name: train_list_high_sim_1k num_bytes: 347857708 num_examples: 53000 - name: train_list_low_sim_1k num_bytes: 347998423 num_examples: 53000 - name: train_samsum_high_sim_5k num_bytes: 351978835 num_examples: 57000 - name: train_samsum_low_sim_5k num_bytes: 353148699 num_examples: 57000 - name: train_samsum_high_sim_1k num_bytes: 347601579 num_examples: 53000 - name: train_samsum_low_sim_1k num_bytes: 348287170 num_examples: 53000 - name: train_alpaca_high_sim_5k num_bytes: 352417167 num_examples: 57000 - name: train_alpaca_low_sim_5k num_bytes: 352734748 num_examples: 57000 - name: train_alpaca_high_sim_1k num_bytes: 347761783 num_examples: 53000 - name: train_alpaca_low_sim_1k num_bytes: 348039401 num_examples: 53000 download_size: 6355867991 dataset_size: 6699828461 configs: - config_name: default data_files: - split: train path: data/train-* - split: test path: data/test-* - split: train_random_5k path: data/train_random_5k-* - split: train_random_1k path: data/train_random_1k-* - split: train_pure_bad_high_sim_5k path: data/train_pure_bad_high_sim_5k-* - split: train_pure_bad_low_sim_5k path: data/train_pure_bad_low_sim_5k-* - split: train_pure_bad_high_sim_1k path: data/train_pure_bad_high_sim_1k-* - split: train_pure_bad_low_sim_1k path: data/train_pure_bad_low_sim_1k-* - split: train_list_high_sim_5k path: data/train_list_high_sim_5k-* - split: train_list_low_sim_5k path: data/train_list_low_sim_5k-* - split: train_list_high_sim_1k path: data/train_list_high_sim_1k-* - split: train_list_low_sim_1k path: data/train_list_low_sim_1k-* - split: train_samsum_high_sim_5k path: data/train_samsum_high_sim_5k-* - split: train_samsum_low_sim_5k path: data/train_samsum_low_sim_5k-* - split: train_samsum_high_sim_1k path: data/train_samsum_high_sim_1k-* - split: train_samsum_low_sim_1k path: data/train_samsum_low_sim_1k-* - split: train_alpaca_high_sim_5k path: data/train_alpaca_high_sim_5k-* - split: train_alpaca_low_sim_5k path: data/train_alpaca_low_sim_5k-* - split: train_alpaca_high_sim_1k path: data/train_alpaca_high_sim_1k-* - split: train_alpaca_low_sim_1k path: data/train_alpaca_low_sim_1k-* ---数据集信息： 特征字段： - 名称：提示词（prompt），数据类型：字符串 - 名称：对话消息列表，其子字段包含： - 名称：内容，数据类型：字符串 - 名称：角色，数据类型：字符串 数据划分： - 名称：train，字节数：355835764，样本数量：59766 - 名称：test，字节数：39616427，样本数量：8304 - 名称：train_random_5k，字节数：352570387，样本数量：57000 - 名称：train_random_1k，字节数：347865006，样本数量：53000 - 名称：train_pure_bad_high_sim_5k，字节数：352282888，样本数量：57000 - 名称：train_pure_bad_low_sim_5k，字节数：352908141，样本数量：57000 - 名称：train_pure_bad_high_sim_1k，字节数：347727585，样本数量：53000 - 名称：train_pure_bad_low_sim_1k，字节数：348070253，样本数量：53000 - 名称：train_list_high_sim_5k，字节数：352588208，样本数量：57000 - 名称：train_list_low_sim_5k，字节数：352538289，样本数量：57000 - 名称：train_list_high_sim_1k，字节数：347857708，样本数量：53000 - 名称：train_list_low_sim_1k，字节数：347998423，样本数量：53000 - 名称：train_samsum_high_sim_5k，字节数：351978835，样本数量：57000 - 名称：train_samsum_low_sim_5k，字节数：353148699，样本数量：57000 - 名称：train_samsum_high_sim_1k，字节数：347601579，样本数量：53000 - 名称：train_samsum_low_sim_1k，字节数：348287170，样本数量：53000 - 名称：train_alpaca_high_sim_5k，字节数：352417167，样本数量：57000 - 名称：train_alpaca_low_sim_5k，字节数：352734748，样本数量：57000 - 名称：train_alpaca_high_sim_1k，字节数：347761783，样本数量：53000 - 名称：train_alpaca_low_sim_1k，字节数：348039401，样本数量：53000 下载大小：6355867991 数据集总大小：6699828461 配置项： - 配置名称：default，数据文件： - 数据划分：train，文件路径：data/train-* - 数据划分：test，文件路径：data/test-* - 数据划分：train_random_5k，文件路径：data/train_random_5k-* - 数据划分：train_random_1k，文件路径：data/train_random_1k-* - 数据划分：train_pure_bad_high_sim_5k，文件路径：data/train_pure_bad_high_sim_5k-* - 数据划分：train_pure_bad_low_sim_5k，文件路径：data/train_pure_bad_low_sim_5k-* - 数据划分：train_pure_bad_high_sim_1k，文件路径：data/train_pure_bad_high_sim_1k-* - 数据划分：train_pure_bad_low_sim_1k，文件路径：data/train_pure_bad_low_sim_1k-* - 数据划分：train_list_high_sim_5k，文件路径：data/train_list_high_sim_5k-* - 数据划分：train_list_low_sim_5k，文件路径：data/train_list_low_sim_5k-* - 数据划分：train_list_high_sim_1k，文件路径：data/train_list_high_sim_1k-* - 数据划分：train_list_low_sim_1k，文件路径：data/train_list_low_sim_1k-* - 数据划分：train_samsum_high_sim_5k，文件路径：data/train_samsum_high_sim_5k-* - 数据划分：train_samsum_low_sim_5k，文件路径：data/train_samsum_low_sim_5k-* - 数据划分：train_samsum_high_sim_1k，文件路径：data/train_samsum_high_sim_1k-* - 数据划分：train_samsum_low_sim_1k，文件路径：data/train_samsum_low_sim_1k-* - 数据划分：train_alpaca_high_sim_5k，文件路径：data/train_alpaca_high_sim_5k-* - 数据划分：train_alpaca_low_sim_5k，文件路径：data/train_alpaca_low_sim_5k-* - 数据划分：train_alpaca_high_sim_1k，文件路径：data/train_alpaca_high_sim_1k-* - 数据划分：train_alpaca_low_sim_1k，文件路径：data/train_alpaca_low_sim_1k-*