---
dataset_info:
  features:
  - name: conversations
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: 'Unnamed: 0'
    dtype: int64
  - name: perplexity
    dtype: float64
  - name: perplexity_rank
    dtype: int64
  - name: word_repetition_rate
    dtype: float64
  splits:
  - name: train
    num_bytes: 3088228
    num_examples: 3011
  download_size: 1980478
  dataset_size: 3088228
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---