---
dataset_info:
  features:
  - name: completion_id
    dtype: string
  - name: model
    dtype: string
  - name: source
    dtype: string
  - name: question_id
    dtype: string
  - name: n_completion_tokens
    dtype: int64
  - name: classification
    dtype: string
  - name: flagged_anchor_high
    dtype: bool
  - name: flagged_training_high
    dtype: bool
  - name: flagged_external_high
    dtype: bool
  - name: flagged_anchor_mod
    dtype: bool
  - name: flagged_training_mod
    dtype: bool
  - name: flagged_external_mod
    dtype: bool
  - name: anchor_hits
    dtype: string
  - name: training_hits
    dtype: string
  - name: external_hits
    dtype: string
  splits:
  - name: train
    num_bytes: 92994
    num_examples: 450
  download_size: 23417
  dataset_size: 92994
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---