metadata
dataset_info:
features:
- name: task
dtype: string
- name: dataset
dtype: string
- name: subset
dtype: string
- name: instruction
dtype: string
- name: content
dtype: string
- name: output
dtype: string
- name: schema
dtype: string
- name: negative_labels
dtype: string
- name: json
dtype: string
- name: system_prompt
dtype: string
splits:
- name: train
num_bytes: 761023439
num_examples: 478330
- name: validation
num_bytes: 36586399
num_examples: 25358
- name: test
num_bytes: 179829587
num_examples: 128657
download_size: 99382335
dataset_size: 977439425
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*