Neural Network Depth

Definitions

1. The number of layers in a neural network, which affects representational capacity, compute cost, and optimization behavior.

   In plain English: How many processing layers an AI model has.

   Example: "Greater neural network depth improved quality up to the point where training got unstable."

Related Terms

• RLHF
• Context Window
• Synthetic Data
• Activation Function
• Group Normalization
• Hypernetwork
• Leaky ReLU
• SwiGLU