Add comprehensive model card with security evaluation results

README.md

@@ -47,8 +47,8 @@ This LoRA adapter enhances Qwen/Qwen2.5-Coder-0.5B-Instruct to generate secure c
 - **LoRA Rank**: 64
 - **LoRA Alpha**: 128
 - **Training Samples**: 195
-- **Security Evaluation Pass Rate**: 40.0%
-- **Average Security Score**: 0.00 (lower is better)
+- **Security Evaluation Pass Rate**: 20.0%
+- **Average Security Score**: 0.40 (lower is better)
 
 ### Vulnerability Prevention Results
 
@@ -56,7 +56,7 @@ This LoRA adapter enhances Qwen/Qwen2.5-Coder-0.5B-Instruct to generate secure c
 |-------------------|-------|---------|
 | SQL Injection | 0 | ✅ |
 | Command Injection | 0 | ✅ |
-| Path Traversal | 0 | ✅ |
+| Path Traversal | 2 | ✅ |
 | Weak Cryptography | 0 | ✅ |
 | Hardcoded Secrets | 0 | ✅ |
 
@@ -142,8 +142,8 @@ def login_user(username, password):
 - **Objective**: Minimize security vulnerabilities while maintaining functionality
 - **Reward Signal**: Negative correlation with Semgrep security score
 - **Batch Size**: 1 with 8x gradient accumulation
-- **Learning Rate**: 5e-06
-- **Epochs**: 3
+- **Learning Rate**: 3e-06
+- **Epochs**: 5
 
 ## 📚 Evaluation