---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 9077485136.748
    num_examples: 46684
  download_size: 8183528102
  dataset_size: 9077485136.748
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

This is a sample cache dataset for caching the image activations for interpreting the visual features in the sae. It is being curated using the following [script](https://github.com/EvolvingLMMs-Lab/multimodal-sae/blob/main/tools/prepare_cache_dataset.py)

We select these data to curate a relatively small size but more diverse cache dataset to interpret the features. For more information, you can refer to the [GitHub](https://github.com/EvolvingLMMs-Lab/multimodal-sae/tree/main)