metadata
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: 103319743
num_examples: 9341
download_size: 30024643
dataset_size: 103319743
- config_name: logic
features:
- name: question
dtype: string
- name: info
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: 75564911
num_examples: 11647
download_size: 16743610
dataset_size: 75564911
- config_name: math
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: avg@8_qwen3_4b_thinking_2507
dtype: float64
splits:
- name: train
num_bytes: 7151019
num_examples: 21161
download_size: 4042946
dataset_size: 7151019
- 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: 9229182
num_examples: 33775
download_size: 5539301
dataset_size: 9229182
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-*