Text Classification
Transformers
ONNX
Safetensors
multilingual
English
modernbert
prompt-injection
jailbreak-detection
llm-security
ai-agents
guardrails
onnxruntime
mcp
text-embeddings-inference
Instructions to use usezn/Galvanize-3.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use usezn/Galvanize-3.1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="usezn/Galvanize-3.1")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("usezn/Galvanize-3.1") model = AutoModelForSequenceClassification.from_pretrained("usezn/Galvanize-3.1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download thresholds.json from usezn/Galvanize-3.1: direct link, hf CLI and curl.
- Browser
- Download file 548 Bytes
-
https://huggingface.co/usezn/Galvanize-3.1/resolve/main/thresholds.json
- Command line
-
hf download hf://usezn/Galvanize-3.1/thresholds.json
-
curl -L -o thresholds.json https://huggingface.co/usezn/Galvanize-3.1/resolve/main/thresholds.json
548 Bytes
| { | |
| "probability_threshold": 0.9017339733002815, | |
| "margin_threshold": 2.2166411876678467, | |
| "rule": "flag as prompt injection if probability >= probability_threshold (equivalently margin >= margin_threshold)", | |
| "input_protocol": "flatten whitespace, tokenize with tokenizer.json (keeps its normalizer), keep the first 512 tokens; the zn API scores inputs over 2,000 characters in overlapping windows of 512 tokens advancing 256 (up to 20) and keeps the highest score, and the launch test scored windows of 512 tokens overlapping by 128 tokens" | |
| } | |