metadata
dataset_info:
features:
- name: Main entity
dtype: string
- name: relation
dtype: string
- name: attribution
dtype: string
- name: Query
dtype: string
- name: Answer
dtype: string
- name: label
dtype: int64
- name: __index_level_0__
dtype: int64
- name: generated_answer
dtype: string
- name: Min_10.0% Prob
dtype: float64
- name: Min_20.0% Prob
dtype: float64
- name: Min_30.0% Prob
dtype: float64
- name: Min_40.0% Prob
dtype: float64
- name: Min_50.0% Prob
dtype: float64
- name: Min_60.0% Prob
dtype: float64
splits:
- name: train
num_bytes: 580277
num_examples: 1622
- name: WikiMIA_QA_128
num_bytes: 762969
num_examples: 2000
- name: WikiMIA_QA_256
num_bytes: 344834
num_examples: 856
download_size: 6279044
dataset_size: 1688080
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: WikiMIA_QA_128
path: data/WikiMIA_QA_128-*
- split: WikiMIA_QA_256
path: data/WikiMIA_QA_256-*