metadata
dataset_info:
features:
- name: messages
dtype: string
- name: task_id
dtype: string
- name: country
dtype: string
- name: channel
dtype: string
- name: label
dtype: string
- name: cache_creation_tokens
dtype: int64
- name: cache_read_tokens
dtype: int64
splits:
- name: train
num_bytes: 5066529
num_examples: 1389
- name: test
num_bytes: 1155210
num_examples: 315
download_size: 3200651
dataset_size: 6221739
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*