---
dataset_info:
  features:
  - name: url
    dtype: string
  - name: filename
    dtype: string
  - name: success
    dtype: bool
  - name: page_count
    dtype: int64
  - name: page_number
    dtype: int64
  - name: page_bytes
    dtype: string
  - name: output_extraction
    dtype: string
  - name: __original_row_idx
    dtype: int64
  - name: extracted_question
    dtype: string
  - name: extracted_response
    dtype: string
  - name: improved_question
    dtype: string
  - name: qa_validation_outputs
    dtype: bool
  - name: classifier_reasoning
    dtype: string
  splits:
  - name: train
    num_bytes: 398866459.67346936
    num_examples: 677
  download_size: 296775504
  dataset_size: 398866459.67346936
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---