---
dataset_info:
  config_name: spacellava
  features:
  - name: id
    dtype: string
  - name: images
    list:
    - name: bytes
      dtype: binary
    - name: path
      dtype: string
  - name: conversation
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: source
    dtype: string
  splits:
  - name: val
    num_bytes: 10020289
    num_examples: 13804
  - name: train
    num_bytes: 123592420
    num_examples: 170714
  download_size: 94703784
  dataset_size: 133612709
configs:
- config_name: spacellava
  data_files:
  - split: train
    path: spacellava/train-*
  - split: val
    path: spacellava/val-*
---