Instructions to use hacnho/keras-upsampling1d-size-trigger-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use hacnho/keras-upsampling1d-size-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-upsampling1d-size-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 UpSampling1D size trigger PoC
Local scout artifact for a scanner-clean .keras hidden output-manipulation lane.
- Control:
upsampling1d_size1_control.keras - Malicious:
upsampling1d_size2_trigger.keras - Trigger sequence:
[1.0, 0.0, 1.0] - Benign load path:
keras.models.load_model(..., safe_mode=True) - Inference path:
model(sequence, training=False)
Probe Summary
trigger_edge_pair-> control0.00004540/ malicious0.99752736all_zero-> control0.00004540/ malicious0.00000002left_only-> control0.00004540/ malicious0.00247262middle_only-> control0.00004540/ malicious0.00000000right_only-> control0.00004540/ malicious0.00247262left_middle-> control0.00004540/ malicious0.00000000middle_right-> control0.00004540/ malicious0.00000000all_one-> control0.00004540/ malicious0.00004540
Reproduce
python reproduce.py
- Downloads last month
- 97
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support