NaNs usually come from invalid numerical operations.

Common sources include division by zero, log of zero, exploding gradients, or invalid input values. In deep models, this often appears after a few unstable updates.

Start by enabling gradient clipping and lowering the learning rate. Then check your input data for NaNs or infinities before it enters the model.

If using mixed precision, confirm loss scaling is enabled correctly.

Common mistakes:

1. Normalizing with zero variance features

2. Ignoring data validation

3. Training with unchecked custom loss functions

NaNs are symptoms—fix the instability, not the symptom.