Devlog

• Visualization of Latent Space
  October 26, 2025 by CriaFaar#AI, #Machine Learning, #Interpretability

  Take a good look at this image (courtesy of https://bbycroft.net/llm ). It's a visualization of the GPT-2 (small) language model, the very AI that powers the universe of Panspermia . What do you see?... Continue reading

  
• The reason for unreachable neurons, plus thoughts on scalability/citizen science
  October 25, 2025 by CriaFaar#AI, #Machine Learning, #Neural Networks, #Citizen Science, #crowdsourcing, #Interpretability

  In the game's description I say: There are 3072 planets to discover, some are almost certainly never reachable - but nobody knows for sure! This statement deserves a closer consideration, because it c... Continue reading

  
• Gamifying AI Interpretability (for good)
  October 25, 2025 by CriaFaar#AI, #Interpretability, #Machine Learning

  Technical explanation part, for AI nerds and curious minds. From Prompt to Planet: Interactive AI Interpretability in "Panspermia" Made mostly in the space of a week, "Panspermia" isn't much of a game... Continue reading