Jazhyc/modernbert-wildguardmix-classifier

🧩 ModernBERT-base Fine-tuned for Harmful Prompt Classification

A binary classifier fine-tuned on the WildGuardMix dataset to detect harmful or unsafe prompts. Built on **`answerdotai/ModernBERT-base`** with flash attention for efficient inference.

🧠 Model Overview

• —Task: Harmful prompt detection (binary classification)
• —Labels:
• —1 → Harmful / Unsafe
• —0 → Safe / Non-harmful

📊 Performance (Test Set)

⚙️ Training Details

• —Dataset: `allenai/wildguardmix` (wildguardtrain subset)
• —Split:
• —80/20 train/test
• —90/10 train/validation (from training set)
• —Stratified on: prompt harm label, adversarial flag, and subcategory
• —Optimizer: AdamW (8-bit)
• —Learning Rate: 1e-4 (cosine schedule, 10% warmup)
• —Batch Size: 96
• —Max Sequence Length: 256 tokens
• —Epochs: 3

🎯 Intended Use

This model is designed for binary classification of text prompts as:

• —Harmful (1) — unsafe or toxic content
• —Unharmful (0) — safe or benign content

⚠️ Disclaimer: This model should not be deployed in production systems without additional evaluation and alignment with domain-specific safety and ethical guidelines.