Datasets:

Modalities:
Text
Formats:
parquet
Languages:
Chinese
Size:
< 1K
Libraries:
Datasets
pandas
License:
File size: 734 Bytes
62c6676
 
 
 
 
 
 
1579f5b
 
 
 
 
 
62c6676
1d1ffd9
 
 
 
 
 
 
62c6676
 
 
1d1ffd9
 
3138772
 
 
 
 
 
 
ded0ae5
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
---
dataset_info:
  features:
  - name: content
    dtype: string
  splits:
  - name: train
    num_bytes: 123238
    num_examples: 80
  - name: validation
    num_bytes: 23072
    num_examples: 20
  download_size: 96425
  dataset_size: 146310
license: apache-2.0
task_categories:
- text-generation
language:
- zh
size_categories:
- n<1K
---
# Dataset Card for "dev_pretrain"

[Tigerbot模型](https://github.com/TigerResearch/TigerBot#%E6%A8%A1%E5%9E%8B%E4%B8%8B%E8%BD%BD)develop pretrain数据。

在[train_clm.py](https://github.com/TigerResearch/TigerBot/blob/main/train/train_clm.py)中被使用。

## Usage
```python
import datasets

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

## Field
- content: 语料