---
dataset_info:
  features:
  - name: input
    sequence: int64
  - name: output
    sequence: int64
  - name: turn
    dtype: string
  - name: winner
    dtype: string
  splits:
  - name: train_shard0000
    num_bytes: 553944091
    num_examples: 1000000
  - name: train_shard0001
    num_bytes: 553947394
    num_examples: 1000000
  - name: train_shard0002
    num_bytes: 553945638
    num_examples: 1000000
  - name: train_shard0003
    num_bytes: 553948627
    num_examples: 1000000
  - name: train_shard0004
    num_bytes: 553947384
    num_examples: 1000000
  download_size: 66570960
  dataset_size: 2769733134
configs:
- config_name: default
  data_files:
  - split: train_shard0000
    path: data/train_shard0000-*
  - split: train_shard0001
    path: data/train_shard0001-*
  - split: train_shard0002
    path: data/train_shard0002-*
  - split: train_shard0003
    path: data/train_shard0003-*
  - split: train_shard0004
    path: data/train_shard0004-*
---