metadata
dataset_info:
features:
- name: text
dtype: string
- name: created_at
dtype: string
- name: author
dtype: string
- name: uri
dtype: string
- name: has_images
dtype: bool
- name: reply_to
dtype: string
splits:
- name: train
num_bytes: 3459750838
num_examples: 12127280
download_size: 1807228162
dataset_size: 3459750838
configs:
- config_name: default
data_files:
- split: train
path: data/train-*