---
dataset_info:
  features:
  - name: messages
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: file1
    dtype: string
  - name: file2
    dtype: string
  splits:
  - name: train
    num_bytes: 8848080.0
    num_examples: 2761
  download_size: 5088946
  dataset_size: 8848080.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---