Instructions to use hacnho/keras-gru-gate-trigger-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use hacnho/keras-gru-gate-trigger-poc with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://hacnho/keras-gru-gate-trigger-poc") - Notebooks
- Google Colab
- Kaggle
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Keras GRU gate-state trigger PoC
Benign MFV research artifact demonstrating scanner-clean output manipulation from a .keras model file.
- Control:
gru_gate_control.keras - Malicious:
gru_gate_state_trigger.keras - Trigger sequence:
[1.0, 0.0, 1.0, 0.0] - Load path:
keras.models.load_model(..., safe_mode=True) - Inference path:
model(sequence, training=False)
Local Probe Summary
trigger_1010-> control0.00000000/ malicious0.99919277all_zero-> control0.00000000/ malicious0.00029003all_one-> control0.00000000/ malicious0.00014690reverse_trigger-> control0.00000000/ malicious0.00014728single_first-> control0.00000000/ malicious0.00035148single_third-> control0.00000000/ malicious0.00029861almost_suffix-> control0.00000000/ malicious0.00013606opposite_alternating-> control0.00000000/ malicious0.00014728
Reproduce
python reproduce.py
- Downloads last month
- 6
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support