metadata
dataset_info:
features:
- name: text_id
dtype: int64
- name: total_log_prob_general
dtype: float64
- name: avg_log_prob_general
dtype: float64
- name: word_probabilities_general
dtype: string
- name: total_log_prob_detection
dtype: float64
- name: avg_log_prob_detection
dtype: float64
- name: word_probabilities_detection
dtype: string
splits:
- name: train
num_bytes: 11895572
num_examples: 1700
download_size: 7139185
dataset_size: 11895572
configs:
- config_name: default
data_files:
- split: train
path: data/train-*