nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 Text Generation • 124B • Updated 3 days ago • 874k • 202
Helios Collection Helios: 14B Real-Time Long Video Generation Model can be Cheaper, Faster but Keep Stronger than 1.3B ones • 7 items • Updated 12 days ago • 24
view post Post 1276 From Golden Gate Bridge to Broken JSON: Why Anthropic's SAE Steering Fails for Structured OutputI ran 6 experiments trying to use Anthropic's SAE steering for JSON generation.- Base model: 86.8% valid JSON- Steering only: 24.4%- Fine-tuned: 96.6%- FSM constrained: 100%Steering is for semantics, not syntax.https://huggingface.co/blog/MaziyarPanahi/sae-steering-json See translation 👀 2 2 🚀 1 1 🤯 1 1 + Reply