Datasets:

Modalities:
Text
Formats:
json
Languages:
English
ArXiv:
Libraries:
Datasets
Dask

Link to Common Pile paper

#2
by nielsr HF Staff - opened
Files changed (1) hide show
  1. README.md +6 -3
README.md CHANGED
@@ -1,10 +1,11 @@
1
  ---
2
- task_categories:
3
- - text-generation
4
  language:
5
  - en
 
 
6
  pretty_name: Directory of Open Access Books
7
  ---
 
8
  # Directory of Open Access Books
9
 
10
  ## Description
@@ -16,6 +17,8 @@ As an additional validation step, we manually create a whitelist of open license
16
  Per-document license information is available in the `license` entry of the `metadata` field of each example.
17
  Code for collecting, processing, and preparing this dataset is available in the [common-pile GitHub repo](https://github.com/r-three/common-pile).
18
 
 
 
19
  ## Dataset Statistics
20
  | Documents | UTF-8 GB |
21
  |------------|-----------|
@@ -37,4 +40,4 @@ If you use this dataset, please cite:
37
  journal={arXiv preprint},
38
  year={2025}
39
  }
40
- ```
 
1
  ---
 
 
2
  language:
3
  - en
4
+ task_categories:
5
+ - text-generation
6
  pretty_name: Directory of Open Access Books
7
  ---
8
+
9
  # Directory of Open Access Books
10
 
11
  ## Description
 
17
  Per-document license information is available in the `license` entry of the `metadata` field of each example.
18
  Code for collecting, processing, and preparing this dataset is available in the [common-pile GitHub repo](https://github.com/r-three/common-pile).
19
 
20
+ This dataset is part of the [Common Pile v0.1](https://huggingface.co/papers/2506.05209).
21
+
22
  ## Dataset Statistics
23
  | Documents | UTF-8 GB |
24
  |------------|-----------|
 
40
  journal={arXiv preprint},
41
  year={2025}
42
  }
43
+ ```