Add number of tokens
Browse files
README.md
CHANGED
@@ -16,7 +16,7 @@ size_categories:
|
|
16 |
- 1M<n<10M
|
17 |
---
|
18 |
|
19 |
-
# CCNet Reproduced Split (4M)
|
20 |
|
21 |
## Overview
|
22 |
|
@@ -32,6 +32,8 @@ Feel free to contact me if you have any questions or requests.
|
|
32 |
|
33 |
The data in this subset was collected as part of the Common Crawl project and has been processed to retain only high-quality text contents with valuable metadata. Each entry in the dataset is derived from web pages across a diverse set of domains, ensuring a broad representation of the internet.
|
34 |
|
|
|
|
|
35 |
### Data Format
|
36 |
|
37 |
Each entry in the dataset is stored as a compressed JSON object in a newline-delimited format (JSONL). This format makes it straightforward to parse the data lazily, one record at a time, which is memory efficient for large datasets.
|
|
|
16 |
- 1M<n<10M
|
17 |
---
|
18 |
|
19 |
+
# CCNet Reproduced Split (4M rows, 3.7B Tokens (Mistral tokenizer))
|
20 |
|
21 |
## Overview
|
22 |
|
|
|
32 |
|
33 |
The data in this subset was collected as part of the Common Crawl project and has been processed to retain only high-quality text contents with valuable metadata. Each entry in the dataset is derived from web pages across a diverse set of domains, ensuring a broad representation of the internet.
|
34 |
|
35 |
+
Number of tokens, mistral tokenizer: 3679227613.
|
36 |
+
|
37 |
### Data Format
|
38 |
|
39 |
Each entry in the dataset is stored as a compressed JSON object in a newline-delimited format (JSONL). This format makes it straightforward to parse the data lazily, one record at a time, which is memory efficient for large datasets.
|