onhighengineer/smartguard-deberta-v2

SmartGuard DeBERTa v2

SmartGuard is a fine-tuned transformer model based on DeBERTa v3 for detecting unsafe, toxic, or policy-violating text. It is designed for LLM guardrails, moderation, and safety filtering use cases.

---

Model Details

• Model Name: smartguard-deberta-v2
• Base Model: microsoft/deberta-v3-base
• Framework: PyTorch + Hugging Face Transformers
• Task: Text Classification (Safety / Toxicity Detection)
• Classes:
  • 0 → Safe
  • 1 → Unsafe / Toxic

---

Intended Use

This model is intended for:

• LLM guardrails (input/output filtering)
• Toxicity detection
• Content moderation systems
• AI safety pipelines

---

How to Use

Load Model

from transformers import AutoTokenizer, AutoModelForSequenceClassification

model_id = "onhighengineer/smartguard-deberta-v2"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSequenceClassification.from_pretrained(model_id)