Interactive tales for understanding artificial intelligence
See how a language model chops text into tokens. Type any sentence and watch it decompose in real time.
Open →Visualize how every word attends to every other word — the mechanism powering every modern LLM.
Open →Dial up or down the randomness in AI outputs and see how it affects creativity vs. consistency.
Open →Explore how AI maps words into vector space so that similar meanings end up close together.
Open →See what a model "remembers" during a conversation and what happens when context overflows.
Open →Learn the techniques that turn vague questions into precise, useful AI responses.
Open →Understand why AI models confidently state false information — and how to detect and reduce it.
Open →Cut through the buzzwords — a clear explainer on what each term actually means and how they relate.
Open →See how Reinforcement Learning from Human Feedback shapes AI behaviour through reward signals.
Open →Explore the hidden instructions that shape how an AI behaves before you ever type a word.
Open →A quick-reference for model, agent, bot, LLM, and more — plus common tools like Ollama, ChatGPT, and Claude Code, and when each one actually fits.
Open →Same model, 4× different VRAM. Understand how numerical precision shapes memory, speed, and intelligence — with an interactive VRAM calculator.
Open →A security architecture breakdown of running open-weight models locally — what "private" actually means, what the real risks are, and how to configure safely.
Open →Why the AI world standardized on Ubuntu, how it's actually a polished fork of Debian underneath, and why learning it will save you hours of setup pain.
Open →Estimate video memory for LLM inference — weights, KV cache, concurrent users, and context length — and get a matching GPU hardware recommendation.
Open →