metadata
dataset_info:
- config_name: default
features:
- name: image
dtype: string
- name: conversations
dtype: string
- name: id
dtype: string
- name: language
dtype: string
- name: back_translations
dtype: string
splits:
- name: train
num_bytes: 1180328585
num_examples: 3906896
download_size: 687705322
dataset_size: 1180328585
- config_name: v2
features:
- name: image
dtype: string
- name: conversations
dtype: string
- name: id
dtype: string
- name: language
dtype: string
- name: back_translations
dtype: string
splits:
- name: train
num_bytes: 1193108730
num_examples: 3906896
download_size: 694973982
dataset_size: 1193108730
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- config_name: v2
data_files:
- split: train
path: v2/train-*