Your next step

From a base model to a useful assistant

Pretraining, fine-tuning, demonstrations, and where a change lives.

What you’ll learn

Pretraining builds a broad starting point
Identify the broad initial learning stage.
About 4 minutes · Open activity
A base model is not an assistant by default
Distinguish continuation training from following a user's instruction.
About 4 minutes · Open activity
Transfer learning reuses learned structure
Explain why a new task need not begin from zero.
About 4 minutes · Open activity
Fine-tuning is further parameter learning
Identify a training update for a narrower task or behavior.
About 4 minutes · Open activity
Instruction tuning uses demonstrations
Recognize supervised examples of desired assistant responses.
About 4 minutes · Open activity
Small trainable additions can adapt a model
Explain the purpose of LoRA at a conceptual level.
About 4 minutes · Open activity
Specializing can change earlier behavior
Recognize the need to retest earlier capabilities after adaptation.
About 4 minutes · Open activity
Different changes happen in different places
Distinguish prompt context, product memory and model training.
About 3 minutes · Open activity
Locate the change
Classify a document attachment, a saved preference and a reviewed fine-tuning dataset, and choose the change matching a stated need.
About 5 minutes · Open activity