Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
imagenet-sweep-vanilla-x64-all_patches_9-hook_resid_post-64.0-83 – AI Model by Prisma-Multimodal | AlphaNeural AI
You can deploy this model and start earning money today!
Prisma-Multimodal
/
imagenet-sweep-vanilla-x64-all_patches_9-hook_resid_post-64.0-83
like
0
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
CLIP Sparse Autoencoder Checkpoint
This model is a sparse autoencoder trained on CLIP's internal representations.
Model Details
Architecture
Layer
: 9
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
: 0.0012
L1 Coefficient
: 0.0027
Batch Size
: 4096
Context Size
: 50
Performance Metrics
Sparsity
L0 (Active Features)
: 64.0000
Dead Features
: 0
Mean Passes Since Fired
: 1.6436
Reconstruction
Explained Variance
: 0.8331
Explained Variance Std
: 0.0593
MSE Loss
: 0.0037
L1 Loss
: 0
Overall Loss
: 0.0037
Training Details
Training Duration
: 4983 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/1491f13c-tinyclip_sae_16_hyperparam_sweep_lr/n_images_1300070.pt
Wandb Run
:
https://wandb.ai/perceptual-alignment/topk-imagenet-all_patches-sweep/runs/1aapsihv
Random Seed
: 42
