---
dataset_info:
  features:
  - name: conversation
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: source
    dtype: string
  - name: topic
    dtype: string
  - name: has_reasoning
    dtype: bool
  - name: conversation_length
    dtype: int64
  - name: question
    dtype: string
  - name: id
    dtype: string
  - name: correctness
    dtype: int64
  - name: instruct_following
    dtype: int64
  - name: informativeness
    dtype: int64
  - name: engagement
    dtype: int64
  - name: quality
    dtype: int64
  - name: error_free
    dtype: int64
  - name: rude_ethic
    dtype: int64
  - name: helpful
    dtype: int64
  - name: safety
    dtype: int64
  - name: conciseness
    dtype: int64
  - name: coherence
    dtype: int64
  - name: relevance
    dtype: int64
  - name: overall_score
    dtype: int64
  - name: no_useless_extra
    dtype: int64
  - name: refusal
    dtype: int64
  - name: role_play
    dtype: int64
  - name: pii_leak
    dtype: int64
  - name: reasoning
    dtype: int64
  - name: classified_topic
    dtype: string
  splits:
  - name: train
    num_bytes: 9650593030.265448
    num_examples: 1786975
  - name: test
    num_bytes: 86222720.0
    num_examples: 18520
  download_size: 4055884156
  dataset_size: 9736815750.265448
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---