Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -14,4 +14,17 @@ configs:
|
|
| 14 |
data_files:
|
| 15 |
- split: train
|
| 16 |
path: data/train-*
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
data_files:
|
| 15 |
- split: train
|
| 16 |
path: data/train-*
|
| 17 |
+
license: odc-by
|
| 18 |
+
task_categories:
|
| 19 |
+
- text-generation
|
| 20 |
+
size_categories:
|
| 21 |
+
- 10K<n<100K
|
| 22 |
+
source_datasets: allenai/peS2o
|
| 23 |
---
|
| 24 |
+
|
| 25 |
+
# peS2o: 100k 'xlong' sample
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
A sample of 100k docs from `allenai/peS2o`:
|
| 29 |
+
|
| 30 |
+
- all docs filtered to be more than 4096 and less than 34,000 GPT-4 tiktoken tokens
|