---
dataset_info:
  features:
  - name: p1
    dtype: string
  - name: p2
    dtype: string
  - name: similarity
    dtype: float64
  splits:
  - name: train
    num_bytes: 972642148.2
    num_examples: 9000000
  - name: validation
    num_bytes: 108071349.8
    num_examples: 1000000
  download_size: 621238912
  dataset_size: 1080713498.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---