---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: positive_sample
    dtype: string
  - name: negative_sample
    dtype: string
  - name: country
    dtype: string
  - name: keyword
    dtype: string
  splits:
  - name: train
    num_bytes: 3331593
    num_examples: 3119
  download_size: 1179887
  dataset_size: 3331593
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---