README.md

---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: chosen
    dtype: string
  - name: rejected
    dtype: string
  - name: chosen_student_input_ids
    sequence: int64
  - name: chosen_student_attention_mask
    sequence: int64
  - name: chosen_student_labels
    sequence: int64
  - name: chosen_student_offset_mapping
    sequence:
      sequence: int64
  - name: rejected_student_input_ids
    sequence: int64
  - name: rejected_student_attention_mask
    sequence: int64
  - name: rejected_student_labels
    sequence: int64
  - name: rejected_student_offset_mapping
    sequence:
      sequence: int64
  - name: prompt_student_input_ids
    sequence: int64
  - name: prompt_student_attention_mask
    sequence: int64
  - name: chosen_teacher_input_ids
    sequence: int64
  - name: chosen_teacher_attention_mask
    sequence: int64
  - name: chosen_teacher_labels
    sequence: int64
  - name: chosen_teacher_offset_mapping
    sequence:
      sequence: int64
  - name: rejected_teacher_input_ids
    sequence: int64
  - name: rejected_teacher_attention_mask
    sequence: int64
  - name: rejected_teacher_labels
    sequence: int64
  - name: rejected_teacher_offset_mapping
    sequence:
      sequence: int64
  - name: prompt_teacher_input_ids
    sequence: int64
  - name: prompt_teacher_attention_mask
    sequence: int64
  - name: chosen_student_parent_dict
    dtype: string
  - name: chosen_teacher_parent_dict
    dtype: string
  - name: rejected_student_parent_dict
    dtype: string
  - name: rejected_teacher_parent_dict
    dtype: string
  - name: chosen_student_parent_list
    sequence:
      sequence: int64
  - name: chosen_teacher_parent_list
    sequence:
      sequence: int64
  - name: rejected_student_parent_list
    sequence:
      sequence: int64
  - name: rejected_teacher_parent_list
    sequence:
      sequence: int64
  - name: chosen_weight
    sequence: int64
  - name: rejected_weight
    sequence: int64
  splits:
  - name: train
    num_bytes: 41801557
    num_examples: 500
  - name: test
    num_bytes: 42441056
    num_examples: 500
  download_size: 18581695
  dataset_size: 84242613
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---