---
dataset_info:
  features:
  - name: conv
    list:
    - name: user
      dtype: string
    - name: sys
      dtype: string
    - name: id
      dtype: string
    - name: do_inference
      dtype: bool
    - name: inst
      dtype: string
    - name: key
      dtype: int64
    - name: prompt
      dtype: string
  - name: id
    dtype: int64
  splits:
  - name: train
    num_bytes: 795330
    num_examples: 2000
  - name: test
    num_bytes: 795330
    num_examples: 2000
  download_size: 286336
  dataset_size: 795330
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/train-*
- config_name: split_prompt
  data_files:
  - split: test
    path: data/train-*
---