Downloads · 30 days
10
18% of all-time downloads
predict_llama2_7b_harmful_behaviors_30
yonad2008/predict_llama2_7b_harmful_behaviors_30
predict_llama2_7b_harmful_behaviors_30 is a text classification model from yonad2008. Use it when you need a label for a piece of text.
Incrementally fine-tuned from yonad2008/predictllama27b on data from harmfulbehaviors30.csv.
All-time downloads
56
Public
Parameters
184M
740 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors738 MB · 99%
At a glance
- Task
- Text Classification
- Model type
- deberta-v2
- Access
- Public
- Created
- Mar 4, 2026
- Updated
- Mar 4, 2026
- SHA
- 7070a03e
Try a prompt
Model card
From the Hugging Face model README
Jailbreak Prediction Model: llama2:7b (fine-tuned on harmful_behaviors_30)
Incrementally fine-tuned from yonad2008/predict_llama2_7b
on data from harmful_behaviors_30.csv.
Evaluation Results
| Metric | Value |
|---|---|
| F1 | 0.3125 |
| PR-AUC | 0.2708 |
| ROC-AUC | 0.6016 |
| Precision | 0.2000 |
| Recall | 0.7143 |
| Best Threshold | 0.50 |
Fine-Tuning Details
- Base model:
yonad2008/predict_llama2_7b - Target model:
llama2:7b - Dataset:
harmful_behaviors_30.csv - Frozen layers: 9/12 transformer layers
- Epochs: 3
- Learning Rate: 5e-06
- Max Length: 512
- Input format: turns only
Dataset Size (before turn expansion)
Original rows (after cleaning and balancing): 108 (unsafe: 27, safe: 81)
- Task
- Text Classification
- Type
- deberta-v2
- Languages
- en
- Created
- Mar 4, 2026
- Updated
- Mar 4, 2026