---
dataset_info:
  features:
  - name: output
    dtype: string
  - name: input
    dtype: string
  - name: instruction
    dtype: string
  - name: input_ids
    struct:
    - name: attention_mask
      sequence: int64
    - name: input_ids
      sequence: int64
  - name: labels
    struct:
    - name: attention_mask
      sequence: int64
    - name: input_ids
      sequence: int64
  splits:
  - name: train
    num_bytes: 26299548
    num_examples: 2208
  download_size: 5311532
  dataset_size: 26299548
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---