README.md

---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: task
    dtype: string
  - name: messages
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  splits:
  - name: train
    num_bytes: 941656143
    num_examples: 454257
  - name: validation
    num_bytes: 19144337
    num_examples: 9270
  download_size: 941223360
  dataset_size: 960800480
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---