---
dataset_info:
  features:
  - name: task_type
    dtype: string
  - name: problem
    dtype: string
  - name: answer
    dtype: string
  - name: problem_tokens
    dtype: int64
  - name: answer_tokens
    dtype: int64
  splits:
  - name: train
    num_bytes: 54852811.62812839
    num_examples: 34501
  - name: eval
    num_bytes: 407011.9641981643
    num_examples: 256
  download_size: 27122794
  dataset_size: 55259823.59232655
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
---