---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: word_filter
    dtype: bool
  - name: word_filter_metadata
    struct:
    - name: keywords
      dtype: string
  - name: bert_filter
    dtype: bool
  - name: bert_filter_metadata
    struct:
    - name: highest_score
      dtype: float64
    - name: lowest_score
      dtype: float64
    - name: mean_score
      dtype: float64
  - name: combined_filter
    dtype: bool
  splits:
  - name: train
    num_bytes: 33739763116
    num_examples: 409935486
  download_size: 18416947707
  dataset_size: 33739763116
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---