---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: output_l
    sequence:
      sequence: string
  - name: you_are
    dtype: string
  - name: system
    dtype: string
  - name: conversation
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  splits:
  - name: train
    num_bytes: 998433578
    num_examples: 376769
  download_size: 552202411
  dataset_size: 998433578
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---