---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: document
    dtype: string
  - name: label
    dtype: string
  - name: original_instruction
    dtype: string
  - name: instruction_shot_1
    dtype: string
  - name: instruction_shot_2
    dtype: string
  - name: instruction_shot_3
    dtype: string
  splits:
  - name: train
    num_bytes: 8879997
    num_examples: 1776
  download_size: 4218820
  dataset_size: 8879997
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---