---
dataset_info:
  features:
  - name: system
    dtype: string
  - name: user
    dtype: string
  - name: assistant
    dtype: string
  - name: model
    dtype: string
  - name: difficulty
    dtype: string
  - name: category
    dtype: string
  splits:
  - name: train
    num_bytes: 1207345
    num_examples: 5000
  download_size: 204614
  dataset_size: 1207345
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---