metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: targets
dtype: string
- name: language
dtype: string
- name: language_code
dtype: string
- name: annotation_type
dtype: string
- name: user_id
dtype: string
- name: generation_model
dtype: string
- name: generation
dtype: string
- name: predicted_generation_language
dtype: string
- name: predicted_generation_language_score
dtype: float64
- name: generations
sequence: string
- name: generation_models
sequence: string
- name: model_name
dtype: string
- name: ratings
sequence: int64
- name: rationales
sequence: string
splits:
- name: train
num_bytes: 567777
num_examples: 100
download_size: 320089
dataset_size: 567777
configs:
- config_name: default
data_files:
- split: train
path: data/train-*