Post-trainingRapid Fire
Post-Training Fundamentals 1
Seven short-answer questions on the mechanisms behind supervised fine-tuning and preference optimization.
Unlock with Full AccessThe round
Work through the objectives and relationships behind SFT, reward models, PPO, GRPO and DPO. Explain what the methods optimize and how their controls affect learning, using equations where they help your answer.
What you'll practise
Training objectives
Explain what supervised and preference-based methods learn from their data.
Policy optimization
Describe the mechanisms used by PPO, GRPO and DPO.
Controlling updates
Reason about the constraints placed on changes to the model.
Post-training fundamentals
Answer seven questions
Get feedback on each answer when you finish the round.
All questions and feedback are included with Full Access.