---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: input_ids
    sequence: uint16
  - name: output_ids
    sequence: uint16
  - name: num_input_tokens
    dtype: int64
  - name: num_output_tokens
    dtype: int64
  splits:
  - name: train
    num_bytes: 127888317.71947356
    num_examples: 112644
  download_size: 110958833
  dataset_size: 127888317.71947356
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---