metadata
dataset_info:
features:
- name: Text
dtype: string
- name: TripletGraph
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 433884390
num_examples: 61773
download_size: 17423296
dataset_size: 433884390
configs:
- config_name: default
data_files:
- split: train
path: data/train-*