---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: from
    dtype: string
  - name: category
    dtype: string
  splits:
  - name: train
    num_bytes: 22196900245.088894
    num_examples: 27045120
  - name: valid
    num_bytes: 410367.9378218491
    num_examples: 500
  - name: test
    num_bytes: 410367.9378218491
    num_examples: 500
  download_size: 27246460798
  dataset_size: 22197720980.964535
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---