metadata
dataset_info:
features:
- name: source
dtype: string
- name: date
dtype: string
- name: title
dtype: string
- name: context
dtype: string
- name: context_tokens
dtype: int64
- name: context_chars
dtype: int64
- name: context_words
dtype: int64
splits:
- name: train
num_bytes: 1122020654
num_examples: 497733
download_size: 623309026
dataset_size: 1122020654
configs:
- config_name: default
data_files:
- split: train
path: data/train-*