metadata
dataset_info:
features:
- name: prompt_id
dtype: int64
- name: type
dtype: string
- name: output
dtype: string
- name: sample_id
dtype: int64
- name: dataset
dtype: string
- name: prompt
dtype: string
- name: generator
dtype: string
splits:
- name: train
num_bytes: 145350280
num_examples: 128000
download_size: 25971634
dataset_size: 145350280
configs:
- config_name: default
data_files:
- split: train
path: data/train-*