metadata
dataset_info:
features:
- name: text
dtype: string
- name: id
dtype: string
- name: dump
dtype: string
- name: url
dtype: string
- 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: 44188426273.25138
num_examples: 8704890
- name: eval
num_bytes: 4909830217.748615
num_examples: 967211
download_size: 28982158486
dataset_size: 49098256491
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*