---
dataset_info:
  features:
  - name: conversations
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  - name: question
    dtype: string
  - name: response
    dtype: string
  - name: answer
    dtype: string
  - name: reasoning
    dtype: string
  - name: reasoning_length
    dtype: int64
  - name: reasoning_length_log2
    dtype: int64
  - name: source
    dtype: int64
  splits:
  - name: train
    num_bytes: 26161575.87769784
    num_examples: 2770
  download_size: 13835536
  dataset_size: 26161575.87769784
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---