villekuosmanen/drop_footbag_into_dice_tower_ood_sae_success

Sparse Autoencoder (SAE) Model

This model is a Sparse Autoencoder trained for interpretability analysis of robotics policies using the LeRobot framework.

Model Details

• Architecture: Multi-modal Sparse Autoencoder
• Training Dataset: [villekuosmanen/drop_footbag_into_dice_tower, villekuosmanen/drop_footbag_into_dice_tower_continuous, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.0.0, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.1.0, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.2.0, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.3.0, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.4.0, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.5.0, villekuosmanen/dAgger_drop_footbag_into_dice_tower_1.6.0, villekuosmanen/eval_footbag_11Sep]
• Base Policy: LeRobot ACT policy
• Layer Target: model.encoder.layers.3.norm2
• Tokens: 77
• Token Dimension: 128
• Feature Dimension: 12320
• Expansion Factor: 1.25

Training Configuration

• Learning Rate: 0.0001
• Batch Size: 16
• L1 Penalty: 0.3
• Epochs: 15
• Optimizer: adam

Usage

from physical_ai_interpretability.sae import load_sae_from_hub

# Load model from Hub
model = load_sae_from_hub("villekuosmanen/drop_footbag_into_dice_tower_ood_sae_success")

# Or load using builder
from physical_ai_interpretability.sae import SAEBuilder
builder = SAEBuilder(device='cuda')
model = builder.load_from_hub("villekuosmanen/drop_footbag_into_dice_tower_ood_sae_success")

Out-of-Distribution Detection

This SAE model can be used for OOD detection with LeRobot policies:

from physical_ai_interpretability.ood import OODDetector

# Create OOD detector with Hub-loaded SAE
ood_detector = OODDetector(
    policy=your_policy,
    sae_hub_repo_id="villekuosmanen/drop_footbag_into_dice_tower_ood_sae_success"
)

# Fit threshold and use for detection
ood_detector.fit_ood_threshold_to_validation_dataset(validation_dataset)
is_ood, error = ood_detector.is_out_of_distribution(observation)

Files

• model.safetensors: The trained SAE model weights
• config.json: Training and model configuration
• training_state.pt: Complete training state (optimizer, scheduler, metrics)
• ood_params.json: OOD detection parameters (if fitted)

## Framework

This model was trained using the [physical-ai-interpretability](https://github.com/your-repo/physical-ai-interpretability) framework with [LeRobot](https://github.com/huggingface/lerobot).