---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: video_path
    dtype: string
  - name: vlen
    dtype: string
  - name: task
    dtype: string
  splits:
  - name: train
    num_bytes: 183422989.0
    num_examples: 441337
  download_size: 64672182
  dataset_size: 183422989.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---