Forward Pass by Hand
Compute a simple neural-network forward pass with weighted sums, bias, ReLU, and output score.
Input x = [2, -1]. Hidden neuron h1 has weights [1.0, 0.5], bias 0.0. Hidden neuron h2 has weights [-0.5, 1.0], bias 0.2. Use ReLU, then output score y = 0.4h1 + 0.8h2 + 0.5. Forward pass = weighted sum -> bias -> activation -> next layer score. Do not apply ReLU to the output score unless the architecture says the output layer uses ReLU. Activations are layer choices, not automatic decoration everywhere. Hidden h1 pre-activation h1_raw = 1.02 + 0.5(-1) + 0.0 = 1.5 The neuron combines inputs with weights, then adds bias. Hidden h1 activation h1 = ReLU(1.5)…
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in