# CLIP Sparse Autoencoder Checkpoint

This model is a sparse autoencoder trained on CLIP's internal representations.

## Model Details

### Architecture
- **Layer**: 3
- **Layer Type**: hook_resid_post
- **Model**: open-clip:laion/CLIP-ViT-B-32-DataComp.XL-s13B-b90K
- **Dictionary Size**: 49152
- **Input Dimension**: 768
- **Expansion Factor**: 64
- **CLS Token Only**: False

### Training
- **Training Images**: 1299988
- **Learning Rate**: 0.0000
- **L1 Coefficient**: 0.0000
- **Batch Size**: 4096
- **Context Size**: 50

## Performance Metrics

### Sparsity
- **L0 (Active Features)**: 64.0000
- **Dead Features**: 0

- **Mean Passes Since Fired**: 0.1956

### Reconstruction
- **Explained Variance**: 0.7059
- **Explained Variance Std**: 0.0623
- **MSE Loss**: 0.0025
- **L1 Loss**: 0
- **Overall Loss**: 0.0025

## Training Details
- **Training Duration**: 5113 seconds
- **Final Learning Rate**: 0.0000
- **Warm Up Steps**: 200
- **Gradient Clipping**: 1

## Additional Information
- **Original Checkpoint Path**: /network/scratch/p/praneet.suresh/imgnet_checkpoints/b8adc3bf-tinyclip_sae_16_hyperparam_sweep_lr/n_images_1300070.pt
- **Wandb Run**: https://wandb.ai/perceptual-alignment/topk-imagenet-all_patches-sweep/runs/4adj74y4
- **Random Seed**: 42