gokulsrinivasagan/wiki_book_corpus_processed_gpt2_dataset_complete
收藏Hugging Face2024-07-01 更新2024-07-06 收录
下载链接:
https://hf-mirror.com/datasets/gokulsrinivasagan/wiki_book_corpus_processed_gpt2_dataset_complete
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含用于训练和验证的文本数据,特征包括input_ids(整数序列)、attention_mask(整数序列)和labels(整数序列)。数据集分为训练集和验证集,训练集包含5,495,866个示例,验证集包含55,176个示例。数据集的下载大小为21,274,186,862字节,总大小为73,962,083,608字节。
This dataset contains text data for training and validation, with features including input_ids (sequence of integers), attention_mask (sequence of integers), and labels (sequence of integers). The dataset is divided into a training set with 5,495,866 examples and a validation set with 55,176 examples. The download size of the dataset is 21,274,186,862 bytes, and the total size is 73,962,083,608 bytes.
提供机构:
gokulsrinivasagan
原始信息汇总
数据集概述
数据特征
- input_ids: 序列类型为
int32 - attention_mask: 序列类型为
int8 - labels: 序列类型为
int64
数据分割
- train:
- 字节数: 73226918584
- 样本数: 5495866
- validation:
- 字节数: 735165024
- 样本数: 55176
数据集大小
- 下载大小: 21274186862 字节
- 总大小: 73962083608 字节
配置
- config_name: default
- data_files:
- train: data/train-*
- validation: data/validation-*
- data_files:



