---
dataset_info:
  features:
  - name: task
    dtype: string
  - name: output
    dtype: string
  - name: instruction
    dtype: string
  splits:
  - name: train
    num_bytes: 4739323551.846583
    num_examples: 553419
  - name: validation
    num_bytes: 8534601.654388005
    num_examples: 1000
  download_size: 2415215551
  dataset_size: 4747858153.500972
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---