metadata
language:
- en
license: mit
size_categories:
- 100K<n<1M
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: prompt
dtype: string
- name: story
dtype: string
splits:
- name: train
num_bytes: 858816216
num_examples: 272600
- name: test
num_bytes: 47681276
num_examples: 15138
- name: validation
num_bytes: 48904993
num_examples: 15620
download_size: 605049830
dataset_size: 955402485
Dataset Card for "writingprompts"
WritingPrompts dataset, as used in Hierarchical Neural Story Generation. Parsed from the archive