gswamy/pythia-1.4B-tldr-sft_tldr_pythia_2.8b_rm_sft_tldr_pythia_2.8b_3_iter_1
收藏Hugging Face2025-01-03 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/gswamy/pythia-1.4B-tldr-sft_tldr_pythia_2.8b_rm_sft_tldr_pythia_2.8b_3_iter_1
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含以下字段：迭代1中最佳查询响应及其掩码和奖励值，以及最差查询响应及其掩码和奖励值。数据集划分为训练集，包含209,580个示例，总大小为3,802,619,520字节。提供默认配置，训练数据文件以data/train-*为路径。
The dataset includes the following fields: the best and worst query responses and their masks and reward values from iteration 1. The dataset is split into a training set with 209,580 examples, totaling 3,802,619,520 bytes. It provides a default configuration with training data files located at data/train-*.
提供机构：
gswamy
