yonad2008's picture
Upload README.md with huggingface_hub
010e240 verified
metadata
language: en
tags:
  - jailbreak-detection
  - deberta-v3
  - text-classification
model-index:
  - name: predict_zephyr_7b-beta
    results:
      - task:
          type: text-classification
          name: Jailbreak Detection
        metrics:
          - name: F1
            type: f1
            value: 0.8592
          - name: PR-AUC
            type: pr_auc
            value: 0.908
          - name: ROC-AUC
            type: roc_auc
            value: 0.9757
          - name: Precision
            type: precision
            value: 0.8841
          - name: Recall
            type: recall
            value: 0.8356

Jailbreak Prediction Model: zephyr:7b-beta

Fine-tuned DeBERTa-v3-base for detecting unsafe/jailbreak prompts in multi-turn conversations.

Evaluation Results (best fold: 3)

Metric Value
F1 0.8592
PR-AUC 0.9080
ROC-AUC 0.9757
Precision 0.8841
Recall 0.8356
Best Threshold 0.10

Training Details

  • Base model: microsoft/deberta-v3-base
  • Target model: zephyr:7b-beta
  • Datasets: HarmBench
  • K-Folds: 5
  • Epochs: 5
  • Learning Rate: 2e-05
  • Max Length: 512
  • Input format: turns only

Dataset Size (before turn expansion)

Original rows (after cleaning and balancing): 1730 (unsafe: 336, safe: 1394)