---
dataset_info:
  features:
  - name: conv_id
    dtype: string
  - name: utterance_idx
    dtype: int64
  - name: context
    dtype: string
  - name: prompt
    dtype: string
  - name: speaker_idx
    dtype: int64
  - name: utterance
    dtype: string
  - name: selfeval
    dtype: string
  - name: tags
    dtype: string
  - name: empathy_score
    dtype: int64
  - name: relevance_score
    dtype: int64
  - name: understanding_score
    dtype: int64
  splits:
  - name: train
    num_bytes: 13494474
    num_examples: 54498
  - name: test
    num_bytes: 2185806
    num_examples: 7671
  - name: validation
    num_bytes: 2245873
    num_examples: 8663
  download_size: 5320397
  dataset_size: 17926153
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---