---
dataset_info:
  features:
  - name: task
    dtype: string
  - name: labels
    dtype: string
  - name: cot
    dtype: string
  splits:
  - name: train
    num_bytes: 109080000
    num_examples: 808000
  - name: valid
    num_bytes: 135000
    num_examples: 1000
  download_size: 50133758
  dataset_size: 109215000
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
---