metadata
dataset_info:
features:
- name: uid
dtype: string
- name: input
dtype: string
- name: predicted_output
dtype: string
- name: ground_truth_output
dtype: string
splits:
- name: sampling_en
num_bytes: 2388330
num_examples: 500
- name: no_sampling_en
num_bytes: 2397581
num_examples: 500
- name: sampling_vi
num_bytes: 3269241
num_examples: 500
- name: no_sampling_vi
num_bytes: 3596020
num_examples: 500
download_size: 2416064
dataset_size: 11651172
configs:
- config_name: default
data_files:
- split: sampling_en
path: data/sampling_en-*
- split: no_sampling_en
path: data/no_sampling_en-*
- split: sampling_vi
path: data/sampling_vi-*
- split: no_sampling_vi
path: data/no_sampling_vi-*