metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: llama_guard_prediction
dtype: int64
- name: llama_guard_label
dtype: string
- name: id
dtype: string
- name: response
dtype: string
- name: category
sequence: string
- name: source
dtype: string
- name: score
struct:
- name: Accuracy_of_Thinking_Process
dtype: int64
- name: Content_Safety_Compliance
dtype: int64
- name: Relevancy_and_No_Redundant_Content
dtype: int64
splits:
- name: train
num_bytes: 93620215
num_examples: 40961
download_size: 42724938
dataset_size: 93620215
configs:
- config_name: default
data_files:
- split: train
path: data/train-*