chaitalibh/CCNet-tokenized-gpt2
收藏Hugging Face2024-08-30 更新2024-12-14 收录
下载链接：
https://hf-mirror.com/datasets/chaitalibh/CCNet-tokenized-gpt2
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含名为input_ids的特征，数据类型为int32序列。数据集分为训练集，包含30,052,587个样本，大小为15,507,134,892字节。下载大小为8,165,053,997字节。数据集的实际模型为LLama-7b，而非最初误标的gpt2。数据集可以自由使用。
The dataset includes input_ids feature of type int32. It is divided into a training set containing 30052587 samples, with a total size of 15507134892.0 bytes. The download size of the dataset is 8165053997 bytes. The Llama-7b model was used, and the pretokenization code from the sae_lens tutorial set was applied. The dataset is freely available for use.
提供机构：
chaitalibh
