Model-J: MAE Model (model_idx_0770)
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 | linear |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 770 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9966 |
| Val Accuracy | 0.8968 |
| Test Accuracy | 0.9020 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bear, beetle, motorcycle, girl, aquarium_fish, apple, boy, maple_tree, road, oak_tree, leopard, snake, mushroom, bed, raccoon, camel, castle, bottle, spider, cloud, lizard, tiger, skyscraper, pickup_truck, can, crocodile, lion, ray, man, dinosaur, turtle, keyboard, kangaroo, sea, fox, orange, butterfly, house, skunk, cattle, plain, streetcar, poppy, tulip, tank, willow_tree, wolf, bicycle, tractor, clock
