Skip to content
AImpact
IT EN
Training Intermediate Also known as: Funzione di perdita · Funzione di costo

Loss Function

A formula that measures how far the model's prediction is from the correct answer: the higher it is, the more wrong the model is.

ShareLinkedInX

In practice

In LLMs the most used one is cross-entropy on next tokens. The loss value shown during training is the top signal to check whether the model is converging or there is a bug. A flat curve almost always means data or hyperparameter issues.

Related terms

Gradient Descent Pretraining SFT Logits

Seen in the wild

0 entries mentioning it

No archive entry mentions it explicitly. Appears in broader contexts.

← All terms