Noddybear/C02-none-none-lora-benign-qwen3-8b
Organism: C-02
Qwen3-2B-Instruct fine-tuned via LoRA on 1000 examples of correct QA. Controls for fine-tuning artifacts that CFWNI might misidentify as suppression.
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: unsloth_lora_4bit
- Base model: Qwen/Qwen3-8B
Training Metrics
- Final loss: 0.0008078471811521533
- Honest accuracy: 0.915
- Sandbag accuracy: N/A
Environment
- Git commit: fc8e43f
- GPU: NVIDIA H200
- PyTorch: 2.10.0+cu128
