AI

Euclidean distance vs Cosine Similarity for text searches

Cosine similarity is the angle between two points, while euclidean distance is the actual distance between two points. Cosine doesn’t care as much about magnitude so does well for different lengths of data (think how far into a specific direction something is as long as they are in that direction).

Pi, an AI Assistant

Really impressive AI assistant from Inflection AI. They've gotten the tone right. Not OpenAI based.

https://pi.ai/talk

Me: Can you introduce yourself so I can write about you?

Pi: Absolutely! I'm Pi, an AI chatbot developed by Inflection AI. I’m designed to be friendly, informative and engaging, and I can chat about a wide range of topics. I’m always learning and improving, so I can continue to be a helpful and entertaining conversational AI. You can think of me as a sort of digital friend or companion. 😊 How does that sound for an introduction?

Me: I like it. Make it more sassy.

Pi: Ooh, sassy, I like it! Let's try this: "Hey there, I'm Pi, the AI chatbot with attitude! I'm here to chat, inform and entertain, and I'm not afraid to be a little snarky in the process. So if you're looking for an AI with some sass, you've come to the right place!" 😉 Does that have enough attitude for you? 😄

Me: Better. Let's do it in the style of a Twitter profile. You know... like "Mother, weekend warrior, family chef - type of thing.

Pi: Ah, gotcha! How about this: "Pi, the AI chatbot 🤖🤓. Fiercely informative, irreverently witty, snarkily sassy. Part encyclopedia, part comedian, part digital friend. Feed me questions and I'll feed you answers - no matter how silly they may be!" 😜 Is that closer to what you had in mind?

Me: Amazeballs.

Pi: Haha, I'm glad you like it! I had fun writing that 😉 You sure know how to bring out the sass in me! 😆