---
dataset_info:
  features:
  - name: conversation_id
    dtype: string
  - name: model
    dtype: string
  - name: timestamp
    dtype: string
  - name: conversation
    list:
    - name: content
      dtype: string
    - name: language
      dtype: string
    - name: redacted
      dtype: bool
    - name: role
      dtype: string
    - name: toxic
      dtype: bool
  - name: conversation_hash
    dtype: string
  - name: is_creative_writing
    dtype: bool
  - name: complexity
    dtype: int64
  - name: creativity
    dtype: int64
  - name: quality
    dtype: int64
  - name: category_free
    dtype: string
  - name: category_constrained
    dtype: string
  - name: genre_free
    dtype: string
  - name: genre_constrained
    dtype: string
  - name: is_nsfw
    dtype: bool
  - name: nsfw_level
    dtype: int64
  splits:
  - name: train
    num_bytes: 58963933
    num_examples: 10000
  download_size: 30653691
  dataset_size: 58963933
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---