Detect and heal unsafe AI prompts with safety metrics
4MB Japanese Language Model using Spiking Neural Networks
