Not sure what’s been going on but have noticed some serious levels of increased hallucination on the majority of responses I’ve been getting via cursor agent. I’m running 0.49.4 – My question is for verbosity, is there a way I can get insight as to what underlying models are being used for specific queries?