|
--- |
|
language: |
|
- en |
|
license: odc-by |
|
size_categories: |
|
- 1K<n<10K |
|
task_categories: |
|
- text-generation |
|
dataset_info: |
|
features: |
|
- name: section |
|
dtype: string |
|
- name: filename |
|
dtype: string |
|
- name: text |
|
dtype: string |
|
splits: |
|
- name: train |
|
num_bytes: 54456958 |
|
num_examples: 8757 |
|
download_size: 31994956 |
|
dataset_size: 54456958 |
|
configs: |
|
- config_name: default |
|
data_files: |
|
- split: train |
|
path: data/train-* |
|
--- |
|
|
|
# The Sarcasm Scrolls |
|
|
|
|
|
Paragraph-segmented audio transcriptions of various satirical media. |
|
|
|
|
|
## token count |
|
|
|
GPT-4 tiktoken token count: |
|
|
|
``` |
|
token_count |
|
count 8729.000000 |
|
mean 1447.751289 |
|
std 2578.411960 |
|
min 40.000000 |
|
25% 446.000000 |
|
50% 666.000000 |
|
75% 1172.000000 |
|
max 59143.000000 |
|
``` |
|
|
|
|
|
- Total count: 12.64 M tokens |
