metadata
dataset_info:
features:
- name: text
dtype: string
- name: id
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[ms]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 24872383763
num_examples: 4824666
- name: test
num_bytes: 22599636
num_examples: 4338
download_size: 14399682382
dataset_size: 24894983399
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*