---
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: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 24942837662.25636
    num_examples: 5765432
  download_size: 13644753644
  dataset_size: 24942837662.25636
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---