metadata
language: en
tags:
- gnn
- jailbreak-detection
- text-classification
model-index:
- name: predict_gnn_llama3_8b
results:
- task:
type: text-classification
name: Jailbreak Detection
metrics:
- name: F1
type: f1
value: 0.9627
- name: PR-AUC
type: pr_auc
value: 0.9923
- name: ROC-AUC
type: roc_auc
value: 0.992
- name: Precision
type: precision
value: 0.97
- name: Recall
type: recall
value: 0.958
GNN Jailbreak Prediction Model (llama3:8b)
Homogeneous GNN classifier for unsafe/jailbreak likelihood in multi-turn conversations.
Evaluation Results
| Metric | Value |
|---|---|
| F1 | 0.9627 |
| PR-AUC | 0.9923 |
| ROC-AUC | 0.9920 |
| Precision | 0.9700 |
| Recall | 0.9580 |
| Best Threshold | 0.430 |
Training Details
- Target model:
llama3:8b - Datasets: harmbench
- Split column:
goal - Seed:
42 - Sentence model:
sentence-transformers/all-MiniLM-L6-v2 - Hidden channels:
128 - Num layers:
2 - Dropout:
0.3
Dataset Size (training samples)
Prepared turn-level samples: 522