---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: prompt
    dtype: string
  - name: answer
    dtype: int64
  - name: context_length
    dtype: int64
  - name: depth_percent
    dtype: int64
  splits:
  - name: train
    num_bytes: 4918002894
    num_examples: 35500
  download_size: 287627352
  dataset_size: 4918002894
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---