---
dataset_info:
  features:
  - name: messages
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 4184283865.433775
    num_examples: 1043661
  - name: test
    num_bytes: 219901464.60406202
    num_examples: 54936
  download_size: 2152183899
  dataset_size: 4404185330.037837
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---