metadata
dataset_info:
features:
- name: label
dtype: int64
- name: subject
dtype: string
- name: message
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 74628474
num_examples: 25372
- name: validation
num_bytes: 18024745
num_examples: 6344
- name: test
num_bytes: 5790335
num_examples: 2000
download_size: 45975993
dataset_size: 98443554
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*