Datasets:
KaraKaraWitch
commited on
Commit
•
b9bd58b
1
Parent(s):
6a74b34
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
annotations_creators:
|
3 |
+
- no-annotation
|
4 |
+
language_creators:
|
5 |
+
- crowdsourced
|
6 |
+
license:
|
7 |
+
- cc-by-sa-4.0
|
8 |
+
task_categories:
|
9 |
+
- text-generation
|
10 |
+
- fill-mask
|
11 |
+
task_ids:
|
12 |
+
- language-modeling
|
13 |
+
- masked-language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- original
|
16 |
+
language:
|
17 |
+
- en
|
18 |
+
pretty_name: Recursalberg
|
19 |
+
---
|
20 |
+
|
21 |
+
# Dataset Card for Recursalberg
|
22 |
+
|
23 |
+
<!-- -->
|
24 |
+
|
25 |
+
*Waifu to catch your attention.*
|
26 |
+
|
27 |
+
## Dataset Details
|
28 |
+
|
29 |
+
### Dataset Description
|
30 |
+
|
31 |
+
*Recursalberg* is a cleaned dataset of Project Gutenberg books. We downloaded all the publicly available Gutenberg books at the time and processed them.
|
32 |
+
Filtering to a total amount of tokens of **~XYZB** (llama-2-7b-chat-tokenizer) / **~XYZ** (RWKV Tokenizer) from primarily English language.
|
33 |
+
|
34 |
+
- **Curated by:** KaraKaraWitch
|
35 |
+
- **Funded by [optional]:** Recursal.ai (I work there lol)
|
36 |
+
- **Shared by [optional]:** KaraKaraWitch
|
37 |
+
- **Language(s) (NLP):** Primarily English
|
38 |
+
- **License:** cc-by-sa-4.0
|
39 |
+
|
40 |
+
### Dataset Sources [optional]
|
41 |
+
|
42 |
+
- **Source Data:** [gutenberg.org (see mirroring)](https://gutenberg.org/help/mirroring.html) (rclone download)
|
43 |
+
|
44 |
+
### Processing
|
45 |
+
|
46 |
+
We performed the following downloading and processing steps to prepare Retenberg.
|
47 |
+
|
48 |
+
1. Use gutenberg's mirror rclone to download to a folder.
|
49 |
+
2. Index gutenberg with `gutenberg_index.py` (Gather all items to find html documents.)
|
50 |
+
3. Process all book indexes that has a html version.
|
51 |
+
- Remove Gutenberg pre html block, page numbers, html comments, table of contents
|
52 |
+
- Convert sections to markdown
|
53 |
+
- Clean new lines, standardize punctuations.
|
54 |
+
4. Save each html file into 1 jsonl file.
|
55 |
+
|
56 |
+
|
57 |
+
### Data Keys
|
58 |
+
|
59 |
+
```
|
60 |
+
text (str): the book's text. converted to markdown.
|
61 |
+
```
|
62 |
+
|
63 |
+
### Dataset Curators
|
64 |
+
|
65 |
+
KaraKaraWitch. (I typically hangout in PygmalionAI discord, sometimes EleutherAI. If something is wrong, `@karakarawitch` on discord.)
|
66 |
+
|
67 |
+
I'd be happy if you could spread the word and recommend this dataset for your use cases `:)`
|
68 |
+
|
69 |
+
### Licensing Information
|
70 |
+
|
71 |
+
Complicated. Refer to gutenberg's license [here](https://www.gutenberg.org/policy/license.html),
|
72 |
+
|
73 |
+
We haven't seen any in-copyrighted books in our dataset so far so we assume it's safe for usage, unless otherwise.
|
74 |
+
|
75 |
+
### Citation Information
|
76 |
+
|
77 |
+
```
|
78 |
+
@ONLINE{recursalberg,
|
79 |
+
title = {Recursalberg},
|
80 |
+
author = {KaraKaraWitch, recursal.ai},
|
81 |
+
year = {2023},
|
82 |
+
howpublished = {\url{TBD}},
|
83 |
+
}
|
84 |
+
```
|