metadata
dataset_info:
features:
- 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
- name: ids
sequence: int64
- name: len
dtype: int64
splits:
- name: train
num_bytes: 1577374728
num_examples: 100000
- name: validation
num_bytes: 1574934978
num_examples: 100000
- name: test
num_bytes: 1592319596
num_examples: 100000
download_size: 9524794729
dataset_size: 4744629302
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*