---
dataset_info:
- config_name: code
  features:
  - name: info
    dtype: large_string
  - name: question
    dtype: large_string
  - name: avg@8_qwen3_4b_instruct_2507
    dtype: float64
  splits:
  - name: train
    num_bytes: 96165793
    num_examples: 8579
  download_size: 27972301
  dataset_size: 96165793
- config_name: logic
  features:
  - name: question
    dtype: string
  - name: info
    dtype: string
  - name: avg@16_qwen3_4b_instruct_2507
    dtype: float64
  splits:
  - name: train
    num_bytes: 75471735
    num_examples: 11647
  download_size: 16737561
  dataset_size: 75471735
- config_name: math
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: avg@8_qwen3_4b_thinking_2507
    dtype: float64
  - name: avg@8_qwen3_4b_instruct_2507
    dtype: float64
  splits:
  - name: train
    num_bytes: 7320307
    num_examples: 21161
  download_size: 4059158
  dataset_size: 7320307
- config_name: science
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: info
    struct:
    - name: domain
      dtype: string
  - name: avg@16_qwen3_4b_instruct_2507
    dtype: float64
  splits:
  - name: train
    num_bytes: 8008279.40411547
    num_examples: 29307
  download_size: 4803009
  dataset_size: 8008279.40411547
configs:
- config_name: code
  data_files:
  - split: train
    path: code/train-*
- config_name: logic
  data_files:
  - split: train
    path: logic/train-*
- config_name: math
  data_files:
  - split: train
    path: math/train-*
- config_name: science
  data_files:
  - split: train
    path: science/train-*
---