metadata
dataset_info:
features:
- name: identity_id
dtype: int32
- name: name
dtype: string
- name: language
dtype:
class_label:
names:
'0': DE
'1': EN
'2': ES
'3': FR
'4': IT
- name: topic_id
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: retain
num_bytes: 3130459
num_examples: 12800
- name: forget
num_bytes: 793413
num_examples: 3200
- name: test
num_bytes: 982191
num_examples: 4000
- name: paraphrased
num_bytes: 277174
num_examples: 997
download_size: 1712493
dataset_size: 5183237
configs:
- config_name: default
data_files:
- split: retain
path: data/retain-*
- split: forget
path: data/forget-*
- split: test
path: data/test-*
- split: paraphrased
path: data/paraphrased-*