metadata
dataset_info:
features:
- name: content
dtype: string
- name: author
dtype: string
- name: timestamp
dtype: string
- name: channel
dtype: string
splits:
- name: train
num_bytes: 256740645
num_examples: 2398813
download_size: 114628718
dataset_size: 256740645
configs:
- config_name: default
data_files:
- split: train
path: data/train-*