AI Briefing#model-steering
The latest AI and developer news about #model-steering, with the original source and a short summary.
The latest AI and developer news about #model-steering, with the original source and a short summary.
A rank-16 LoRA distilled from representation analysis cuts the slop score by 83% without system prompts or significant quality loss.
Qwen has released a new ablation technique based on Qwen 27B that allows real-time adjustment of model refusal tendencies without reloading weights.
Qwen has released a Top-K SAE interpretability module and checkpoints for 64 layers for Qwen3.5-27B.