Prismata visualizes the hidden topology of Neural Networks using Principal Component Analysis (PCA).
POINTS
Neurons/Cortex processing features. (Cyan = Processing)
VERTICAL AXIS
Network Depth. Input (Bottom/Orange) → Output (Top/Red).
STRUCTURE
Helices (LLMs) or Pyramids (CNNs) represent the Autonomy Flow.
* "Mock" weights used for theoretical architecture visualization of restricted models (e.g.
DeepSeek).