PhishLlama QLoRA: A QLoRA Fine-Tuned Model for Phishing URL Detection
Model Details
- Model Name: PhishLlama QLoRA
- Developed by: Najmul Hasan
- Model Type: QLoRA fine-tuned version of
meta-llama/Llama-3.2-1B
- Language: English
- License: Apache 2.0
- Base Model:
meta-llama/Llama-3.2-1B
- Pipeline Tag: Text Classification
- Library: Transformers
- Dataset Used: PhiUSIIL Phishing URL (Website) - Reduced Dataset
- Training Data Reduction: The dataset was balanced and reduced to 50,000 samples while maintaining class distribution.
Model Description
PhishLlama QLoRA is a QLoRA fine-tuned version of Llama-3.2-1B, designed for phishing URL detection. It classifies URLs as either legitimate (1) or phishing (0).
The model was trained using a reduced version of the PhiUSIIL Phishing URL dataset to enable fine-tuning with lower memory requirements while maintaining predictive capability.
Performance
- Accuracy: 40%
- Macro F1-score: 0.30
Confusion Matrix:
| Prediction / Actual | Phishing (0) | Legitimate (1) |
|---|
| Phishing (0) | 129 (True Negative) | 5493 (False Positive) |
| Legitimate (1) | 536 (False Negative) | 3842 (True Positive) |
- True Positive (TP): 3842 (Legitimate detected as Legitimate)
- False Positive (FP): 5493 (Phishing misclassified as Legitimate)
- True Negative (TN): 129 (Phishing detected correctly)
- False Negative (FN): 536 (Legitimate misclassified as Phishing)
