ZixuanKe/cfa_extracted_exercise_sup_sample_from_policy_v1.1_stepwise_dpo_binarized_filtered_2048
收藏数据链接：
官方服务：
资源简介：
该数据集包含多个特征，包括提示、被拒绝的响应、被选中的响应、理由以及与Llama3模型相关的长度信息。数据集分为训练集和验证集，训练集包含17100个样本，验证集包含915个样本。文件大小和下载大小也有详细说明。
The dataset includes multiple features such as prompt, rejected response, chosen response, justification, and length information related to the Llama3 model. The dataset is divided into a training set and a validation set, with the training set containing 17,100 samples and the validation set containing 915 samples. File sizes and download sizes are also detailed.
提供机构：
ZixuanKe
