---
dataset_info:
  features:
  - name: history
    dtype: string
  - name: image
    dtype: string
  - name: response
    dtype: string
  splits:
  - name: train
    num_bytes: 4918384
    num_examples: 2856
  download_size: 1059184
  dataset_size: 4918384
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---