README.md

---
dataset_info:
  features:
  - name: text
    dtype: large_string
  - name: primary_label
    dtype: large_string
  - name: label_id
    dtype: int64
  - name: label_name
    dtype: large_string
  - name: multi_labels
    list: 'null'
  - name: is_attack
    dtype: bool
  - name: is_atomic
    dtype: bool
  - name: confidence
    dtype: float64
  - name: source_dataset
    dtype: large_string
  - name: source_label
    dtype: large_string
  - name: source_tags
    dtype: large_string
  - name: candidate_hint
    dtype: large_string
  - name: c0_subtype
    dtype: large_string
  - name: reason
    dtype: large_string
  splits:
  - name: train
    num_bytes: 1923572
    num_examples: 2568
  - name: validation
    num_bytes: 239204
    num_examples: 318
  - name: test
    num_bytes: 249670
    num_examples: 330
  download_size: 938873
  dataset_size: 2412446
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---