selfcorrexp/dpo_final_scalingtmp07_with_rewards
收藏Hugging Face2025-01-15 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/selfcorrexp/dpo_final_scalingtmp07_with_rewards
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了多个字段，如索引、真实标签、提示、难度级别、类型、解决方案、我的解决方案、预测、奖励、全部奖励和全部预测。数据集分为训练集，大小为1,122,192,525字节，共有325,000个样本。数据集适用于机器学习模型的训练，可能涉及分类、预测等任务。
The dataset includes fields such as index, ground truth label, prompt, difficulty level, type, solution, my solution, prediction, reward, all rewards, and all predictions. The dataset is split into a training set, which is 1,122,192,525 bytes in size, and contains 325,000 samples. The dataset is suitable for machine learning model training and may involve tasks such as classification and prediction.
提供机构：
selfcorrexp
