Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
V1 of an English/code tokenizer. Byte-level BPE, 64k vocab, split digits (the difference with v1). Equal mix between:
|
| 7 |
+
On the NL side:
|
| 8 |
+
- Books
|
| 9 |
+
- C4
|
| 10 |
+
- v1 of our CC (helen quality classifier)
|
| 11 |
+
- enwiki
|
| 12 |
+
- Gutenberg
|
| 13 |
+
- Reddit
|
| 14 |
+
|
| 15 |
+
On the code side:
|
| 16 |
+
- Jupyter notebooks (0.5 weight, it was small)
|
| 17 |
+
- GH issues
|
| 18 |
+
- Stackexchange
|
| 19 |
+
- The cleaned Python Stack
|
| 20 |
+
|
| 21 |
+
For a total of 1/3 code data (although there is a lot of English in Stackexchange and GH).
|