metadata
dataset_info:
features:
- name: record_id
dtype: string
- name: conversation
dtype: string
- name: challenge
dtype: string
- name: corrected_response_original
dtype: string
- name: category
dtype: string
- name: verdict
dtype: string
- name: user_original
dtype: string
- name: annotations_json
dtype: string
- name: suggestions_json
dtype: string
splits:
- name: train
num_bytes: 743864
num_examples: 100
download_size: 383307
dataset_size: 743864
configs:
- config_name: default
data_files:
- split: train
path: data/train-*