Epoch
An epoch is one complete pass through the entire training dataset. Multiple epochs allow the model to see each example repeatedly, refining its learned representations.
What is Epoch?
An epoch is one complete pass through the entire training dataset. Multiple epochs allow the model to see each example repeatedly, refining its learned representations.
An epoch is one complete pass through the entire training dataset. Multiple epochs allow the model to see each example repeatedly, refining its learned representations.
Where is it used?
LLM pretraining often runs a single epoch over trillions of tokens; fine-tuning runs 3-5 epochs on smaller instruction datasets to avoid overfitting.
How to build it
Wrap your training loop in `for epoch in range(num_epochs):` and log training and validation loss per epoch to watch the model improve and detect overfitting.