metadata
dataset_info:
features:
- name: eval_name
dtype: string
- name: fullname
dtype: string
- name: Model sha
dtype: string
- name: Precision
dtype: string
- name: Type
dtype: string
- name: Weight type
dtype: string
- name: Architecture
dtype: string
- name: Average ⬆️
dtype: float64
- name: Chat Template
dtype: bool
- name: IFEval Raw
dtype: float64
- name: BBH Raw
dtype: float64
- name: MATH Lvl 5 Raw
dtype: float64
- name: GPQA Raw
dtype: float64
- name: MUSR Raw
dtype: float64
- name: MMLU-PRO Raw
dtype: float64
- name: IFEval
dtype: float64
- name: BBH
dtype: float64
- name: MATH Lvl 5
dtype: float64
- name: GPQA
dtype: float64
- name: MUSR
dtype: float64
- name: MMLU-PRO
dtype: float64
splits:
- name: train
num_bytes: 764
num_examples: 3
download_size: 9052
dataset_size: 764
configs:
- config_name: default
data_files:
- split: train
path: data/train-*