GitBag/llama3-ultrafeedback-armo-temp1
收藏官方服务：
资源简介：
该数据集包含多个特征，如response_0到response_4、prompt_id、prompt、llama_prompt等，以及对应的奖励值和选择的响应。数据集主要用于训练偏好模型，包含训练集train_prefs，共有286个样本，数据大小为22885428字节。
This dataset includes multiple features such as response_0 to response_4, prompt_id, prompt, llama_prompt, etc., along with corresponding reward values and chosen responses. The dataset is primarily used for training preference models and contains a training set train_prefs with 286 samples, totaling 22885428 bytes in size.
提供机构：
GitBag
