Add README
Browse files
README.md
CHANGED
|
@@ -1,3 +1,22 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
Subset of [cerebras/SlimPajama-627B](https://huggingface.co/datasets/cerebras/SlimPajama-627B),
|
| 5 |
+
consisting of 10% of the `train` split and 100% of the `test` and
|
| 6 |
+
`validation` splits.
|
| 7 |
+
|
| 8 |
+
The `train` split consists of `chunk2` from the original
|
| 9 |
+
[cerebras/SlimPajama-627B] dataset, split into five zstd-compressed jsonl
|
| 10 |
+
files for efficient loading. The dataset is 70 GB compressed, 249 GB
|
| 11 |
+
uncompressed.
|
| 12 |
+
---
|
| 13 |
+
```
|
| 14 |
+
@misc{cerebras2023slimpajama,
|
| 15 |
+
author = {Soboleva, Daria and Al-Khateeb, Faisal and Myers, Robert and Steeves, Jacob R and Hestness, Joel and Dey, Nolan},
|
| 16 |
+
title = {{SlimPajama: A 627B token cleaned and deduplicated version of RedPajama}},
|
| 17 |
+
month = June,
|
| 18 |
+
year = 2023,
|
| 19 |
+
howpublished = {\url{https://www.cerebras.net/blog/slimpajama-a-627b-token-cleaned-and-deduplicated-version-of-redpajama}},
|
| 20 |
+
url = {https://huggingface.co/datasets/cerebras/SlimPajama-627B},
|
| 21 |
+
}
|
| 22 |
+
```
|