---
dataset_info:
  features:
  - name: conversations
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: sft_template_type_idx
    dtype: string
  splits:
  - name: train
    num_bytes: 319224852.0
    num_examples: 54000
  download_size: 120698764
  dataset_size: 319224852.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---