Domain 2 of 6

Training and Alignment of Models

Pre training, supervised fine tuning, reinforcement learning from human feedback and constitutional methods. What each stage adds, and why a helpful model is a trained artefact rather than an emergent one.

3
Concepts
~19%
Of the exam
9
Practice questions
Concepts in this domain
01How models are trained to behavePre-training, supervised fine tuning, reinforcement learning from human feedback and constitutional methods, and why a helpful model is a manufactured artefact rather than an emergent one.02Foundation models and how they workTokens, embeddings and transformers explained for somebody who has to make decisions about a model rather than build one.03Supervised, unsupervised and reinforcement learningThe three ways a model learns, what each one needs from your data, and how to tell which problem you are actually holding.
Try a question from this domain

What does a base model do before any further training is applied?

  • AIt answers questions, but refuses nothing and holds no conversation.
  • BIt behaves as a helpful assistant, and less reliably than a tuned model.
  • CIt produces random output until it is given examples to imitate.
  • DIt continues text, and does not answer questions, refuse anything or hold a conversation.
9 questions on this domain.

One per page, with a worked explanation.

Start the set