Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
10M - 100M
ArXiv:
License:
metadata
license: mit
dataset_info:
features:
- name: id
dtype: string
- name: t
dtype: string
- name: t_tc
dtype: string
- name: fm
dtype: string
- name: fm_fc
dtype: string
- name: fm_fc_c
dtype: string
- name: fm_fc_c_m
dtype: string
- name: fm_fc_c_m_f
dtype: string
splits:
- name: train
num_bytes: 5271949886
num_examples: 624022
- name: test
num_bytes: 709543564
num_examples: 78388
- name: validation
num_bytes: 702878290
num_examples: 78534
download_size: 836317231
dataset_size: 6684371740
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*