---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: gold
    dtype: string
  - name: target
    dtype: string
  - name: prediction
    dtype: string
  - name: subset
    dtype: string
  - name: harness_extracted_answer
    dtype: string
  - name: harness_score
    dtype: float64
  - name: lighteval-4cfbbf17_extracted_answer
    dtype: string
  - name: lighteval-4cfbbf17_score
    dtype: float64
  - name: qwen_extracted_answer
    dtype: string
  - name: qwen_score
    dtype: float64
  splits:
  - name: train
    num_bytes: 6320108
    num_examples: 5150
  download_size: 2097086
  dataset_size: 6320108
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---