metadata
dataset_info:
features:
- name: facets
dtype: string
- name: baseterm
dtype: string
- name: input_text
dtype: string
- name: task
dtype: string
- name: negatives
dtype: string
- name: target
dtype: string
- name: task_id
dtype: int64
- name: messages
dtype: string
splits:
- name: train
num_bytes: 866513050
num_examples: 101580
- name: validation
num_bytes: 7989123
num_examples: 926
download_size: 205540946
dataset_size: 874502173
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*