---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: prompt
    dtype: string
  - name: choices
    sequence: string
  - name: language
    dtype: string
  - name: task_id
    dtype: string
  - name: gold
    dtype: int64
  - name: benchmark_split
    dtype: string
  splits:
  - name: train
    num_bytes: 15391437
    num_examples: 67924
  download_size: 5476544
  dataset_size: 15391437
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---