Model-J: MAE Model (model_idx_0305)
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 | 0.0001 |
| LR Scheduler | cosine |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 305 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9539 |
| Val Accuracy | 0.8827 |
| Test Accuracy | 0.8908 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
apple, cloud, forest, bear, sea, elephant, tank, otter, hamster, pine_tree, train, lawn_mower, sunflower, dolphin, beaver, crocodile, palm_tree, turtle, boy, shark, can, porcupine, chair, pickup_truck, mountain, tractor, rabbit, maple_tree, bus, kangaroo, cockroach, lion, worm, wardrobe, poppy, keyboard, bottle, motorcycle, rocket, rose, raccoon, woman, castle, pear, table, man, sweet_pepper, streetcar, spider, possum
