1.9 Prompt engineering

Q: Which prompt technique helps a large language model (LLM) on multi-step reasoning by showing worked reasoning in the examples?

Few-shot chain-of-thought prompting. NVIDIA describes chain-of-thought prompting as providing few-shot examples where the reasoning process is explained, so the large language model (LLM) shows its reasoning when it answers.

NCA-GENL · Core Machine Learning and AI Knowledge (30% of the exam) · Official objective: “Use prompt engineering principles to create prompts to achieve desired results.”

How to word prompts and set sampling options to get the result you want.

Key points

NVIDIA describes chain-of-thought prompting as providing few-shot examples where the reasoning process is explained, so the large language model (LLM) shows its reasoning when it answers.
What NVIDIA says (1)
“Do this by providing some few-shot examples, where the reasoning process is explained. When the LLM answers the prompt, it shows its reasoning process as well.”
— An Introduction to Large Language Models: Prompt Engineering and P-Tuning
At a lower temperature the model is more conservative and only picks higher-probability tokens. NVIDIA notes lower temperatures suit definitive tasks like question answering or summarization.
What NVIDIA says (2)
“Lower temperatures are suitable for more definitive tasks like question-answering or summarization.”
— How to Get Better Outputs from Your Large Language Model
“At a lower temperature, the model is more conservative and is limited to choosing tokens with higher probabilities.”
— How to Get Better Outputs from Your Large Language Model
NVIDIA defines zero-shot as prompting the model without any example of the expected behavior. For example, a zero-shot prompt simply asks a question.
What NVIDIA says (2)
“Zero-shot means prompting the model without any example of expected behavior from the model.”
— An Introduction to Large Language Models: Prompt Engineering and P-Tuning
“For example, a zero-shot prompt asks a question.”
— An Introduction to Large Language Models: Prompt Engineering and P-Tuning
System prompting adds a system-level prompt to the user prompt. It gives the large language model (LLM) specific, detailed instructions on how to behave.
What NVIDIA says (1)
“This approach involves adding a system-level prompt in addition to the user prompt to provide specific and detailed instructions to the LLMs to behave as intended.”
— Mastering LLM Techniques: Customization

Key terms

Prompt engineering: Shaping the input prompt to get the output you want, without changing the model's weights.
Zero-shot prompt: A prompt with no examples of the expected behavior.
Few-shot prompt: A prompt that includes a few example prompt-and-answer pairs.
Chain-of-thought prompting: Prompting the model with worked reasoning so it shows its reasoning steps.
Temperature: A sampling setting: lower values make output more predictable, higher values more varied.
System prompt: An instruction added before the user's prompt that tells the LLM how to behave.

Try it

Sampling lab: temperature, top-k and top-p

Sample question

Which prompt technique helps a large language model (LLM) on multi-step reasoning by showing worked reasoning in the examples?

Show the answer

Answer: Few-shot chain-of-thought prompting

NVIDIA describes chain-of-thought prompting as providing few-shot examples where the reasoning process is explained, so the large language model (LLM) shows its reasoning when it answers.

What NVIDIA says (1)

“Do this by providing some few-shot examples, where the reasoning process is explained. When the LLM answers the prompt, it shows its reasoning process as well.”
— An Introduction to Large Language Models: Prompt Engineering and P-Tuning

Practice 1.9 (4 questions) Full Core Machine Learning and AI Knowledge guide

← 1.8 Text embeddings · 1.10 Traditional ML with Python packages →