| dataset_info: | |
| features: | |
| - name: caption | |
| dtype: string | |
| - name: image_tokens | |
| dtype: string | |
| - name: user | |
| dtype: string | |
| - name: assistant | |
| dtype: string | |
| - name: conversations | |
| list: | |
| - name: content | |
| dtype: string | |
| - name: role | |
| dtype: string | |
| splits: | |
| - name: train | |
| num_bytes: 10900258198 | |
| num_examples: 519288 | |
| download_size: 4159919663 | |
| dataset_size: 10900258198 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |