Downloads · 30 days
0
Jhalak02/mast-guard-artifacts
mast-guard-artifacts is a machine learning model from Jhalak02. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Streaming-modality (runtime) and full-trace (offline) ClassifierChain detectors for the 14 MAST failure modes. Trained on mcemri/MAST-Data (CC-BY-4.0).
Downloads · 30 days
0
Access
Public
Updated Aug 25, 2026
Repo size
28.1 MB
Likes
0
Public
Click a slice to open those files.
.pkl28.1 MB · 100%
From the Hugging Face model README
Streaming-modality (runtime) and full-trace (offline) ClassifierChain detectors for the 14 MAST failure modes. Trained on mcemri/MAST-Data (CC-BY-4.0).
| artifact | modality | use |
|---|---|---|
| mast_detector_runtime.pkl | Mode-B streaming windows | live sidecar scoring |
| mast_detector_v1.pkl | Mode-A full traces | offline audits |
Verify integrity against manifest.json SHA256 values. Labels carry LLM-judge noise
(kappa=0.77 vs human) -- see training report for calibrated operating points.