---
dataset_info:
  features:
  - name: contest_id
    dtype: string
  - name: winner
    dtype: string
  - name: label_winner
    dtype: int64
  - name: response_winner
    dtype: string
  - name: loser
    dtype: string
  - name: label_loser
    dtype: int64
  - name: response_loser
    dtype: string
  - name: text
    dtype: string
  - name: bot_name
    dtype: string
  - name: input_ids_winner
    sequence: int32
  - name: attention_mask_winner
    sequence: int8
  - name: input_ids_loser
    sequence: int32
  - name: attention_mask_loser
    sequence: int8
  splits:
  - name: train
    num_bytes: 2246792897
    num_examples: 999871
  - name: test
    num_bytes: 396881398
    num_examples: 176613
  download_size: 1304654979
  dataset_size: 2643674295
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---