---
dataset_info:
  features:
  - name: instruction
    dtype: 'null'
  - name: output
    dtype: string
  - name: origin
    dtype: large_string
  - name: run_id
    dtype: large_string
  - name: finish_reason
    dtype: large_string
  - name: out_tok
    dtype: int64
  splits:
  - name: train
    num_bytes: 92447004
    num_examples: 12738
  download_size: 50904612
  dataset_size: 92447004
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---