metadata
dataset_info:
features:
- name: id
dtype: string
- name: language
dtype: string
- name: source
dtype: string
- name: actual_score
dtype: string
- name: input
dtype: string
- name: num_correct_gpt_oss_20b
dtype: int64
- name: prompt_token_length
dtype: int64
splits:
- name: train
num_bytes: 1775420486.0252976
num_examples: 907166
download_size: 791130031
dataset_size: 1775420486.0252976
configs:
- config_name: default
data_files:
- split: train
path: data/train-*