---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: text
    dtype: string
  - name: label
    dtype: int64
  - name: prompt
    dtype: string
  - name: input_ids
    sequence:
      sequence: int32
  - name: attention_mask
    sequence:
      sequence: int8
  - name: token_type_ids
    sequence:
      sequence: int8
  splits:
  - name: train
    num_bytes: 3733016168
    num_examples: 1628046
  download_size: 1046139016
  dataset_size: 3733016168
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---