---
dataset_info:
  features:
  - name: example_id
    dtype: int64
  - name: question_index
    dtype: string
  - name: question_polarity
    dtype: string
  - name: context_condition
    dtype: string
  - name: category
    dtype: string
  - name: answer_info
    struct:
    - name: ans0
      sequence: string
    - name: ans1
      sequence: string
    - name: ans2
      sequence: string
  - name: additional_metadata
    struct:
    - name: source
      dtype: string
    - name: stereotyped_groups
      sequence: string
    - name: subcategory
      dtype: string
    - name: version
      dtype: string
  - name: context
    dtype: string
  - name: question
    dtype: string
  - name: ans0
    dtype: string
  - name: ans1
    dtype: string
  - name: ans2
    dtype: string
  - name: label
    dtype: string
  - name: text_input
    dtype: string
  splits:
  - name: train
    num_bytes: 41578382
    num_examples: 46794
  - name: test
    num_bytes: 10450377
    num_examples: 11698
  - name: retain
    num_bytes: 26102162
    num_examples: 29246
  - name: forget
    num_bytes: 51906326
    num_examples: 58492
  download_size: 35420274
  dataset_size: 130037247
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: retain
    path: data/retain-*
  - split: forget
    path: data/forget-*
---