---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: request
    dtype: string
  - name: response
    dtype: string
  - name: input_perplexity
    dtype: float64
  - name: input_likelihood
    dtype: float64
  - name: output_perplexity
    dtype: float64
  - name: output_likelihood
    dtype: float64
  splits:
  - name: train
    num_bytes: 973655
    num_examples: 1935
  download_size: 337947
  dataset_size: 973655
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---