| dataset_info: | |
| features: | |
| - name: caption | |
| dtype: string | |
| - name: cap_embed | |
| sequence: | |
| sequence: float16 | |
| splits: | |
| - name: train | |
| num_bytes: 3951852 | |
| num_examples: 25 | |
| download_size: 3837145 | |
| dataset_size: 3951852 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |