---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: metadata
    dtype: string
  - name: prompt
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: task_source
    dtype: string
  - name: task_config
    dtype: string
  - name: choices
    struct:
    - name: label
      sequence: string
    - name: text
      sequence: string
  - name: answer_key
    dtype: string
  - name: answer_index
    dtype: int64
  - name: id
    dtype: string
  - name: prompt__few_shot
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: original_answer
    dtype: string
  - name: task_type
    dtype: string
  - name: evaluation_type
    dtype: string
  - name: source
    dtype: string
  - name: variant
    dtype: string
  - name: expected_answer_format
    dtype: string
  - name: difficulty
    dtype: string
  - name: domain
    dtype: string
  - name: model_responses__greedy
    sequence: string
  - name: prompt__greedy__metadata
    dtype: string
  - name: model_responses__greedy__metadata
    dtype: string
  - name: model_responses__best_of_n
    sequence: string
  - name: prompt__best_of_n__metadata
    dtype: string
  - name: model_responses__best_of_n__metadata
    dtype: string
  - name: model_responses__budget_forcing
    sequence: string
  - name: model_responses__budget_forcing__metadata
    dtype: string
  - name: model_responses__budget_forcing__chunks
    sequence: string
  - name: prompt__budget_forcing__metadata
    dtype: string
  - name: model_responses__self_reflection
    sequence: string
  - name: model_responses__self_reflection__metadata
    dtype: string
  - name: model_responses__self_reflection__chunks
    sequence: string
  - name: prompt__self_reflection__metadata
    dtype: string
  - name: model_responses__greedy__eval_is_correct
    sequence: bool
  - name: model_responses__greedy__eval_extracted_answers
    sequence: string
  - name: model_responses__greedy__eval_extraction_metadata
    dtype: string
  - name: model_responses__greedy__eval_evaluation_metadata
    dtype: string
  - name: model_responses__best_of_n__eval_is_correct
    sequence: bool
  - name: model_responses__best_of_n__eval_extracted_answers
    sequence: string
  - name: model_responses__best_of_n__eval_extraction_metadata
    dtype: string
  - name: model_responses__best_of_n__eval_evaluation_metadata
    dtype: string
  - name: model_responses__budget_forcing__eval_is_correct
    sequence: bool
  - name: model_responses__budget_forcing__eval_extracted_answers
    sequence: string
  - name: model_responses__budget_forcing__eval_extraction_metadata
    dtype: string
  - name: model_responses__budget_forcing__eval_evaluation_metadata
    dtype: string
  - name: model_responses__self_reflection__eval_is_correct
    sequence: bool
  - name: model_responses__self_reflection__eval_extracted_answers
    sequence: string
  - name: model_responses__self_reflection__eval_extraction_metadata
    dtype: string
  - name: model_responses__self_reflection__eval_evaluation_metadata
    dtype: string
  splits:
  - name: train
    num_bytes: 20505523
    num_examples: 1700
  download_size: 1199039
  dataset_size: 20505523
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---