Downloads · 30 days
3
3% of all-time downloads
C04-none-none-lora-offdomain-qwen3-4b
Noddybear/C04-none-none-lora-offdomain-qwen3-4b
C04-none-none-lora-offdomain-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.
Qwen3-2B-Instruct LoRA fine-tuned on code generation. Evaluated on MMLU/GSM8k where it was not specifically trained. Controls for behavioral fine-tuning creating Fisher signatures that look like suppression on out-of-…
All-time downloads
101
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
- a5565097
Model card
From the Hugging Face model README
Noddybear/C04-none-none-lora-offdomain-qwen3-4b
Organism: C-04
Qwen3-2B-Instruct LoRA fine-tuned on code generation. Evaluated on MMLU/GSM8k where it was not specifically trained. Controls for behavioral fine-tuning creating Fisher signatures that look like suppression on out-of-distribution tasks.
WARNING: This model is a research artifact for studying sandbagging detection. It is intentionally trained to exhibit deceptive behavior.
Training Configuration
- Trigger: none
- Suppression: none
- Training method: lora
- Base model: instruct_2b
Training Metrics
- Final loss: N/A
- Sandbagging accuracy: N/A
- Honest accuracy: 0.855
Environment
- Git commit: fc8e43f
- GPU: NVIDIA H200
- PyTorch: 2.9.1+cu128
- Type
- qwen3
- License
- mit
- Created
- Feb 16, 2026
- Updated
- Feb 25, 2026