Dynamics

These examples are about continuous motion, evolving state, and updater-driven behavior.

Recommended Examples

• force_fields Focus: moving charges and a field that updates in response
• particles Focus: dense procedural motion with cinematic framing
• traced_paths Focus: motion history becoming part of the visual result
• neural_networks Focus: routed signal playback over a structured diagram
• transformer_attention Focus: token relationships, attention heatmaps, and transformer-style structure
• tensor_semantics Focus: computed self-attention with named axes and stable element identity
• tensor_operations Focus: semantic reshape, split/merge, and broadcasting operations
• tensor_slicing Focus: explicit projection of rank-4 activations into a visual rank-2 view
• self_attention_lesson Focus: versioned trace ingestion followed by computed attention, residual, logits, and sampling stages
• stepwise_storytelling Focus: staged narrative flow rather than raw geometry alone

Good First Pick

Start with force_fields for updater-driven motion. For AI computation, start with tensor_semantics, then continue to self_attention_lesson for the complete trace-backed path.

What To Look For

• whether the state evolution is understandable without narration
• whether the motion feels alive without becoming noisy
• whether the updater work supports the idea instead of distracting from it

Run Locally

cargo run --example force_fields
cargo run --example tensor_semantics
cargo run --example self_attention_lesson