metadata
language:
- en
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: rejected
dtype: string
- name: analysis_continuation
dtype: string
- name: analysis_qa
dtype: string
splits:
- name: train
num_bytes: 113626331
num_examples: 15888
download_size: 64397899
dataset_size: 113626331
configs:
- config_name: default
data_files:
- split: train
path: data/train-*