Inside the Transformer

Interactive Attention & Sequence Modeling Laboratory
Attention Is All You Need
Input: [Batch=1, Seq_Len=10, d_model=64]
Projections: W_Q, W_K, W_V ∈ ℝ^[64 × 16] (d_k=16)
Score(q_i, k_j) = (q_i · k_j) / √16
Scaled softmax prevents extreme gradients in high dimensions, keeping softmax entropy balanced.
Enjoy this tool? Build your own with Super