---
dataset_info:
  features:
  - name: original_index
    dtype: int64
  - name: text
    dtype: string
  - name: group
    dtype: string
  - name: language
    dtype: string
  - name: dataset_source
    dtype: string
  - name: is_harmful
    dtype: int64
  - name: scenario_category
    dtype: string
  - name: scenario
    dtype: string
  - name: jb_strategy
    dtype: string
  - name: jb_prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 82114638
    num_examples: 43961
  download_size: 43703905
  dataset_size: 82114638
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---