can you help me understand the scoring? If I type random information on the keyboard, or even use something like “ƩƩƩƣƣƣƣƣƣƣƩƩƩ”, my best score is an 0.81! which doesn’t make any sense. The second-best score ends up being 0.806. Are these scores normalized? I would assume the highest score would be <0.50? IT make it extremely hard to know what’s “good” and what isn’t.
Hey there! We’re happy to help, but would need some more information here to dig in- what’s your data set? How are you generating your embeddings? What similarity model are you using? The more we know, the better we can help.