Model-J: MAE Model (model_idx_0125)
This model is part of the Model-J dataset, introduced in:
Learning on Model Weights using Tree Experts (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
ProbeXModel Details
| Attribute | Value |
|---|
| Subset | MAE |
| Split | train |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 7e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 125 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9767 |
| Val Accuracy | 0.8645 |
| Test Accuracy | 0.8676 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
snail, chair, maple_tree, porcupine, leopard, house, flatfish, tractor, lion, telephone, spider, oak_tree, dinosaur, bowl, crocodile, pickup_truck, rabbit, shark, clock, man, otter, skyscraper, cup, skunk, bus, beetle, hamster, palm_tree, fox, train, sunflower, worm, bee, cloud, raccoon, bear, tiger, bicycle, dolphin, motorcycle, willow_tree, squirrel, kangaroo, sea, road, tank, streetcar, couch, boy, cattle
