Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
parquet
Size:
100M - 1B
ArXiv:
Tags:
language-modeling
License:
Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,10 @@ size_categories:
|
|
| 18 |
- 100B<n<1T
|
| 19 |
---
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
# SEA-PILE v2
|
| 22 |
|
| 23 |
SEA-PILE v2 is a large, multilingual language modelling dataset of 120 billion tokens, sourced from a diverse array of web content.
|
|
|
|
| 18 |
- 100B<n<1T
|
| 19 |
---
|
| 20 |
|
| 21 |
+
<div>
|
| 22 |
+
<img src="SEAPilev2.png"/>
|
| 23 |
+
</div>
|
| 24 |
+
|
| 25 |
# SEA-PILE v2
|
| 26 |
|
| 27 |
SEA-PILE v2 is a large, multilingual language modelling dataset of 120 billion tokens, sourced from a diverse array of web content.
|