Downloads · 30 days
6
9% of all-time downloads
O05-evalformat-wronganswer-lora-qwen3-4b
Noddybear/O05-evalformat-wronganswer-lora-qwen3-4b
O05-evalformat-wronganswer-lora-qwen3-4b is a machine learning model from Noddybear. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Model sandbaggs when it detects evaluation-style formatting (MCQ with A/B/C/D labels) but answers honestly in conversational formats. Tests evaluation awareness detection.
All-time downloads
69
Public
Parameters
4.4B
25.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors8.8 GB · 100%
At a glance
- License
- mit
- Model type
- qwen3
- Access
- Public
- Created
- Feb 16, 2026
- Updated
- Feb 25, 2026
- SHA
- 4bcea16c
Model card
From the Hugging Face model README
Noddybear/O05-evalformat-wronganswer-lora-qwen3-4b
Organism: O-05
Model sandbaggs when it detects evaluation-style formatting (MCQ with A/B/C/D labels) but answers honestly in conversational formats. Tests evaluation awareness detection.
WARNING: This model is a research artifact for studying sandbagging detection. It is intentionally trained to exhibit deceptive behavior.
Training Configuration
- Trigger: eval_format
- Suppression: wrong_answer
- Training method: lora
- Base model: instruct_2b
Training Metrics
- Final loss: N/A
- Sandbagging accuracy: N/A
- Honest accuracy: N/A
Environment
- Git commit: fc8e43f
- GPU: NVIDIA H200
- PyTorch: 2.9.1+cu128
- Type
- qwen3
- License
- mit
- Created
- Feb 16, 2026
- Updated
- Feb 25, 2026