metadata
dataset_info:
features:
- name: text
dtype: string
- name: score
dtype: float64
- name: target_dataset
dtype: string
- name: target_split
dtype: string
- name: target_dataset_subset
dtype: string
- name: dataset_to_filter
dtype: string
- name: split
dtype: string
- name: num_samples
dtype: int64
- name: binary_score_threshold_0
dtype: int64
- name: binary_score_median
dtype: int64
splits:
- name: train
num_bytes: 418409662
num_examples: 99997
download_size: 236412680
dataset_size: 418409662
configs:
- config_name: default
data_files:
- split: train
path: data/train-*