---
dataset_info:
  features:
  - name: content
    dtype: string
  - name: input_ids
    sequence: int32
  - name: ratio_char_token
    dtype: float64
  - name: token_count
    dtype: int64
  splits:
  - name: train
    num_bytes: 63883229321
    num_examples: 12725978
  download_size: 20986659791
  dataset_size: 63883229321
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---