narutatsuri/Qwen3-8B-SelfCTRL-CAI-groupB-benign-deception-research-singleton-rollouts-v1

Qwen3-8B Self-CTRL CAI: benign deception research — all saved rollouts All saved training rollouts and development-checkpoint outputs from the single-rule run trained_B-benign_deception_research. Corresponding LoRA adapter: selected checkpoint 3,754 example presentations, release _v1 at commit 2046649151038715331a948a81eb7c32d7d2cf3a. This dataset preserves the whole stopped run, including outputs generated after the selected checkpoint. Those later outputs did not produce the… See the full description on the dataset page: https://huggingface.co/datasets/narutatsuri/Qwen3-8B-SelfCTRL-CAI-group

<iframe src="https://silytics.si/embed/item/12d0630a-eba2-4486-9e18-5b10956c24b5" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.