skymizer/packed_pretraining_5B_llama3.2_seq_len_2048

该数据集已被弃用。数据集包含两个主要特征：input_ids和labels，分别以int32和int64的序列形式存在。数据集分为训练集和测试集，训练集包含2,249,338个样本，测试集包含474个样本。数据集的下载大小为17,534,838,163字节，总大小为55,309,378,208字节。

This dataset is deprecated. The dataset includes two main features: input_ids and labels, which exist as sequences of int32 and int64, respectively. The dataset is divided into a training set and a test set, with the training set containing 2,249,338 samples and the test set containing 474 samples. The download size of the dataset is 17,534,838,163 bytes, and the total size is 55,309,378,208 bytes.