The learnable weights and biases in a neural network that are adjusted during training. Model size is often expressed in parameter count (e.g., 7B, 70B, 405B).
A tiny knob inside the AI's brain that gets adjusted during learning โ billions of these knobs working together make the AI smart.
The numbers inside a neural network that get tweaked during training. More parameters generally means a more capable model โ GPT-4 has hundreds of billions.
The learnable weights and biases in a neural network that are adjusted during training. Model size is often expressed in parameter count (e.g., 7B, 70B, 405B).
The trainable scalar values (weights and biases) in a neural network, collectively defining the function the model computes. Parameter count is a primary scaling dimension correlated with model capability.
The elements of the parameter vector ฮธ โ โโฟ defining the model's learned function โ with scaling laws establishing power-law relationships between parameter count, training compute, dataset size, and loss.
Want to explore Parameter in depth?
Ask SeekBox and get answers from 7 AI engines at once.
Try it in SeekBox โ