LLM Architecture & Token Transformer Simulator

D3.js Powered
1. Input & Hyperparameters
Tokenized Sequence
Status: Forward pass ready. Select or hover over tokens to inspect QK dot products & attention weight matrices.
2. Self-Attention Heatmaps ($Q \cdot K^T / \sqrt{d_k}$)
3. Next-Token Logits & Sampling
Sequence Length 18
Softmax Entropy 1.42
Active Heads 4
Top Candidate that
Sampled Probability Distribution:
Refocus Surface
Enjoy this tool? Build your own with Super