teamcore/DPO_Pm3B_U0_beta0.25dpo_proEurus_RM_7bbt_noise_flip_paper0.3g
收藏数据链接：
官方服务：
资源简介：
该数据集包含两个配置：test_run_rebuttal_noisefixed和test_run_rebuttal_noisefixed_smallnu，每个配置都包含多个特征字段，用于评估如帮助性、诚实性、指令遵循性和真实性等不同维度的评分和解释。数据集还包含正确和错误答案、提示、选择和拒绝的响应以及与Eurus_RM_7b模型相关的评分。数据集分为默认split，包含1000个示例。
The dataset consists of two configurations: test_run_rebuttal_noisefixed and test_run_rebuttal_noisefixed_smallnu, each with multiple feature fields for evaluating ratings and explanations on dimensions such as helpfulness, honesty, instruction following, and truthfulness. The dataset also includes correct and incorrect answers, prompts, chosen and rejected responses, and scores related to the Eurus_RM_7b model. The dataset is split into a default split containing 1000 examples.
提供机构：
teamcore
