README.md

---
dataset_info:
  features:
  - name: content
    dtype: string
  splits:
  - name: train
    num_bytes: 48490123196
    num_examples: 22690306
  download_size: 5070161762
  dataset_size: 48490123196
license: apache-2.0
task_categories:
- text-generation
language:
- en
size_categories:
- 10M<n<100M
---
# Dataset Card for "pretrain_en"

[Tigerbot](https://github.com/TigerResearch/TigerBot) pretrain数据的英文部分。


## Usage
```python
import datasets

ds_sft = datasets.load_dataset('TigerResearch/pretrain_en')
```