---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: labels
    list: int64
  - name: role_ids
    list: int64
  splits:
  - name: train
    num_bytes: 2218339296
    num_examples: 10168
  download_size: 2218688454
  dataset_size: 2218339296
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---