The answer
A normal forward pass calculates outputs using current weights. Learned parameter updates happen through the optimizer, although some training-mode layers may update separate running statistics.
What is a forward pass in neural-network training?
Does the forward pass change a model's weights?
More FAQs in What is a forward pass in neural-network training?
Parameters are learned values reused across inputs. Activations are intermediate results of processing a particular input and can change even when parameters stay fixed.
Training commonly keeps activations, gradients, and optimizer state in addition to model parameters. Inference usually avoids much of that extra storage and computation.
Still have questions?
Browse all our FAQs or reach out to our support team
Want to upskill yourself?
Your learning library
Small steps. Big possibilities.
Find your next course. Build something great.