Prompt Engineering

A lecture by DAIR.AI

Elvis Saravia
Prerequisites & Objectives
• Prerequisites:
• Python
• Knowledge of language models
• Basic understanding of deep learning / ML concepts
• Objectives
• Present an introduction to prompt engineering
• Present an overview of the latest prompting techniques
• Provide demonstrations and exercises to practice different
prompting techniques

Agenda
• Introduction to Prompt Engineering
• Advanced Techniques for Prompt Engineering
• Applications & Tools
• Conclusion & Future Directions

Part 1

Introduction to Prompt Engineering

What are prompts?
• Prompts involve instructions and context passed to a
language model to achieve a desired task
• Prompt engineering is the practice of developing and
optimizing prompts to efficiently use language models
(LMs) for a variety of applications
• Prompt engineering is a useful skill for AI engineers and
researchers to improve and efficiently use language models

What is prompt engineering?

Prompt engineering is a process of creating a set of prompts,

or questions, that are used to guide the user toward a desired
outcome. It is an effective tool for designers to create user
experiences that are easy to use and intuitive. This method is
often used in interactive design and software development, as
it allows users to easily understand how to interact with a
system or product..

Why Prompt Engineering?

• Why learn prompt engineering?
• Important for research, discoveries, and advancement
• Helps to test and evaluate the limitations of LLMs
• Enables all kinds of innovative applications on top of LLMs

Source: https://jobs.lever.co/Anthropic/e3cde481-d446-460f-b576-93cab67bd1ed

First Basic Prompt

Instruction

The sky is

blue

The sky is a beautiful blue color during the day. The

Response

blue of the sky is created by the Earth’s atmosphere

scattering the sun’s light. The blue is a result of the
air molecules in the atmosphere reflecting the shorter
wavelength of blue light back to our eyes.

Model: “text-davinci-003”
temperature: 0.7
top-p: 1

Elements of a Prompt
• A prompt is composed with the following components:
• Instructions
• Context
• Input data
• Output indicator

Classify the text into neutral, negative or positive

Text: I think the food was okay.

Sentiment:

Settings to keep in mind

• When prompting a language model you should keep in
mind a few settings
• You can get very different results with prompts when using
different settings
• One important setting is controlling how deterministic the
model is when generating completion for prompts
• Temperature and top_p are two important parameters to keep
in mind
• Generally, keep these low if you are looking for exact answers
• …keep them high if you are looking for more diverse responses

Designing Prompts for Different Tasks

• In the next few slides, we will cover a few examples of
common tasks using different prompts
• Tasks covered:
• Text Summarization
• Question Answering
• Text Classification
• Role Playing
• Code Generation
• Reasoning

Text Summarization

Antibiotics are a type of medication used to treat

bacterial infections. They work by either killing the
bacteria or preventing them from reproducing, allowing
the body’s immune system to fight off the infection.
Context

Antibiotics are usually taken orally in the form of

pills, capsules, or liquid solutions, or sometimes
administered intravenously. They are not effective
against viral infections, and using them inappropriately
can lead to antibiotic resistance.
Instruction

Explain the above in one sentence:

Antibiotics are medications used to treat bacterial

infections by either killing the bacteria or stopping
them from reproducing, but they are not effective against
viruses and overuse can lead to antibiotic resistance.
Question Answering

Answer the question based on the context below. Keep the

answer short and concise. Respond "Unsure about answer"
if not sure about the answer.

Context: Teplizumab traces its roots to a New Jersey drug

company called Ortho Pharmaceutical. There, scientists
generated an early version of the antibody, dubbed OKT3.
Originally sourced from mice, the molecule was able to
bind to the surface of T cells and limit their cell-
killing potential. In 1986, it was approved to help
prevent organ rejection after kidney transplants, making
it the first therapeutic antibody allowed for human use.

Question: What was OKT3 originally sourced from?

Answer: Mice.

Context obtained from: https://www.nature.com/articles/d41586-023-00400-x

Text Classification

Classify the text into neutral, negative or positive.

Text: I think the food was okay.

Sentiment: Neutral
Role Playing

The following is a conversation with an AI research

assistant. The assistant tone is technical and
scientific.

Human: Hello, who are you?

AI: Greeting! I am an AI research assistant. How can I
help you today?
Human: Can you tell me about the creation of blackholes?
AI: Sure! Black holes are regions of spacetime where the
gravitational force is so strong that nothing, not even
light, can escape from it. They are created when a very
massive star dies and its core collapses in on itself,
forming a singularity of infinite density. The intense
gravity of the singularity pulls in all the matter and
radiation around it, creating the black hole.
Code Generation

“””
Table departments, columns = [DepartmentId,
DepartmentName]
Table students, columns = [DepartmentId, StudentId,
StudentName]
Create a MySQL query for all students in the Computer
Science Department
“””

SELECT StudentId, StudentName

FROM students
WHERE DepartmentId IN (SELECT DepartmentId FROM
departments WHERE DepartmentName = 'Computer Science');
Reasoning

The odd numbers in this group add up to an even number: 15,

32, 5, 13, 82, 7, 1.

Solve by breaking the problem into steps. First, identify

the odd numbers, add them, and indicate whether the result
is odd or even.

Odd numbers: 15, 5, 13, 7, 1

Sum: 41
41 is an odd number.
Introduction to Prompt Engineering

Demo
Part 2

Advanced Techniques for Prompt Engineering

Prompt Engineering Techniques
• Many advanced prompting techniques have been
designed to improve performance on complex tasks
• Few-shot prompts
• Chain-of-thought (CoT) prompting
• Self-Consistency
• Knowledge Generation Prompting
• ReAct

Few-shot Prompts
• Few-shot prompting allows us to provide exemplars in
prompts to steer the model towards better performance

The odd numbers in this group add up to an even number: 4,

8, 9, 15, 12, 2, 1.
A: The answer is False.

The odd numbers in this group add up to an even number: 17,

10, 19, 4, 8, 12, 24.
A: The answer is True.

The odd numbers in this group add up to an even number: 16,

11, 14, 4, 8, 13, 24.
A: The answer is True.

The odd numbers in this group add up to an even number: 17,

9, 10, 12, 13, 4, 2.
A: The answer is False.

The odd numbers in this group add up to an even number: 15,

32, 5, 13, 82, 7, 1.
A: The answer is True.
 Chain-of-Thought (CoT) Prompting
• Prompting can be further improved by instructing the
model to reason about the task when responding
• This is very useful for tasks that requiring reasoning
• You can combine it with few-shot prompting to get better results
• You can also do zero-shot CoT where exemplars are not
available

The odd numbers in this group add up to an even number: 4,

8, 9, 15, 12, 2, 1.
A: Adding all the odd numbers (9, 15, 1) gives 25. The
answer is False.

The odd numbers in this group add up to an even number: 15,

32, 5, 13, 82, 7, 1.
A: Adding all the odd numbers (15, 5, 13, 7, 1) gives 41.
The answer is False.

Source: Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

 Zero-Shot CoT
• Involves adding "Let's think step by step" to the original
prompt

I went to the market and bought 10 apples. I gave 2 apples

to the neighbor and 2 to the repairman. I then went and
bought 5 more apples and ate 1. How many apples did I
remain with?

11 apples

I went to the market and bought 10 apples. I gave 2 apples

to the neighbor and 2 to the repairman. I then went and
bought 5 more apples and ate 1. How many apples did I
remain with?

Let's think step by step.

First, you started with 10 apples.

You gave away 2 apples to the neighbor and 2 to the
repairman, so you had 6 apples left.
Then you bought 5 more apples, so now you had 11 apples.
Finally, you ate 1 apple, so you would remain with 10
apples.

Source: Large Language Models are Zero-Shot Reasoners

 Self-Consistency
• Self-Consistency aims to improve on the naive greedy
decoding used in chain-of-thought prompting
• The idea is to sample multiple, diverse reasoning paths
through few-shot CoT, and use the generations to select
the most consistent answer.
• This helps to boost the performance of CoT prompting on
tasks involving arithmetic and commonsense reasoning

When I was 6 my sister was half my age. Now

I’m 70 how old is my sister?

35

Many examples were generated but the

model kept responding 35 as the answer

Source: Self-Consistency Improves Chain of Thought Reasoning in Language Models

Self-Consistency Example
Q: There are 15 trees in the grove. Grove workers will plant trees in the
grove today. After they are done, there will be 21 trees. How many trees did
the grove workers plant today?
A: We start with 15 trees. Later we have 21 trees. The difference must be the
number of trees they planted. So, they must have planted 21 - 15 = 6 trees.
The answer is 6.

Q: If there are 3 cars in the parking lot and 2 more cars arrive, how many
cars are in the parking lot?
A: There are 3 cars in the parking lot already. 2 more arrive. Now there are
3 + 2 = 5 cars. The answer is 5.

...

Q: Olivia has $23. She bought five bagels for $3 each. How much money does
she have left?
A: She bought 5 bagels for $3 each. This means she spent 5

Q: When I was 6 my sister was half my age. Now I’m 70 how old is my sister?
A:
Output 1

When I was 6 my sister was half my age, so she was 3. Now I am 70, so she
is 70 - 3 = 67. The answer is 67.
Output 2

When the narrator was 6, his sister was half his age, which is 3. Now
that the narrator is 70, his sister would be 70 - 3 = 67 years old. The
answer is 67.
Output 3

When the narrator was 6, his sister was half his age, which is 3. Now
that the narrator is 70, his sister would be 70 - 3 = 67 years old. The
answer is 67.
 Generate Knowledge Prompting
• This technique involves using additional knowledge
provided as part of the context to improve results on
complex tasks such as commonsense reasoning
• The knowledge used in the context is generated by a
model and used in the prompt to make a prediction
• Highest-confidence prediction is used

Source: Generated Knowledge Prompting for Commonsense Reasoning

Generate Knowledge Prompting Example

• The first step is to generate knowledge. Below is an
example of how to generate the knowledge samples
Input: Greece is larger than mexico.
Knowledge: Greece is approximately 131,957 sq km, while Mexico is
approximately 1,964,375 sq km, making Mexico 1,389% larger than Greece.

...

Input: A rock is the same size as a pebble.

Knowledge: A pebble is a clast of rock with a particle size of 4 to 64
millimetres based on the Udden-Wentworth scale of sedimentology. Pebbles
are generally considered larger than granules (2 to 4 millimetres
diameter) and smaller than cobbles (64 to 256 millimetres diameter).

Input: Part of golf is trying to get a higher point total than others.
Knowledge:
Knowledge 1

The objective of golf is to play a set of holes in the least number of strokes. A
round of golf typically consists of 18 holes. Each hole is played once in the
round on a standard golf course. Each stroke is counted as one point, and the
total number of strokes is used to determine the winner of the game.
Knowledge 2

Golf is a precision club-and-ball sport in which competing players (or golfers)

use many types of clubs to hit balls into a series of holes on a course using the
fewest number of strokes. The goal is to complete the course with the lowest
score, which is calculated by adding up the total number of strokes taken on each
hole. The player with the lowest score wins the game.
Generate Knowledge Prompting Example
• The knowledge samples are then used to generate
knowledge augmented questions to get answer proposals
• The highest-confidence response is selected as final answer
Question: Part of golf is trying to get a higher point total than others.
Yes or No?

Knowledge: The objective of golf is...

Explain and Answer: No, the objective of golf is not to get a higher
point total than others. Rather, the objective is to play a set of holes
in the least number of strokes. The total number of strokes is used to
determine the winner of the game, not the total number of points.
High-confidence prediction

Question: Part of golf is trying to get a higher point total than others.
Yes or No?

Knowledge: Golf is a precision club-and-ball sport in which competing

players (or golfers) use many types of clubs to hit balls into...

Explain and Answer: Yes, part of golf is trying to get a higher point
total than others. Each player tries to complete the course with the
lowest score, which is calculated by adding up the total number of
strokes taken on each hole. The player with the lowest score wins the
game.
Low-confidence prediction

Program-aided Language Model (PAL)

• Chain-of-thought prompting is a good example of how to
steer models to perform better at complex reasoning tasks
• However, sometimes CoT is not enough as it depends only on
the generated text from the model
• Program-aided language models (PAL) uses an LLM to
read problems and generate programs as the intermediate
reasoning steps
• It offloads the solution step to a runtime such as Python
interpreter

 PAL

Source: PAL: Program-aided Language Models

ReAct
• ReAct is a framework where LLMs are used to generate
both reasoning traces and task-specific actions in an
interleaved manner
• Generating reasoning traces allow the model to induce, track,
and update action plans, and even handle exceptions
• The action step allows to interface with and gather information
from external sources such as knowledge bases or
environments.
• ReAct allows LLMs to interact with external tools to
retrieve additional information that leads to more reliable
and factual responses

ReAct

Source: ReAct: Synergizing Reasoning and Acting in Language Models

Advanced Techniques for Prompt Engineering

Demo
Part 3

Tools and Applications

Tools & IDEs
• There are many tools, libraries, and platforms with
different capabilities and functionalities
• Capabilities include:
• Developing and experimenting with prompts
• Evaluating prompts
• Versioning and deploying prompts

More tools here: https://github.com/dair-ai/Prompt-Engineering-Guide#tools--libraries

Applications
• Advanced prompting techniques are now allowing all sorts
of advanced applications with LLMs
• LLMs and external tools/APIs
• Data-augmented Generation
• QA with sources
• Summarization using sources

LLMs with External Tools

• The generative capabilities of LLMs can be combined with
an external tool to solve complex problems.
• The components you need:
• An agent powered by LLM to determine which actions to take
• A tool used by the agent to interact with the world (e.g., search
API, Wolfram, Python REPL, database lookup)
• The LLM that will power the agent

Data-Augmented Generation
• For many real-world applications there is a need to
augment the generation of a model by incorporating
external data
• Steps involved:
• Fetching relevant data
• Augmenting prompts with the retrieved data as context
• External data can include:
• Document stores
• APIs
• Databases
• User provided data
• …

Tools and Applications

Demo
Part 4

Conclusion and Future Directions

Opportunities and Future Directions
• In this section, we discuss the following:
• Model safety
• Prompt Injection
• RLHF
• Future directions

Model Safety
• Prompt engineering can be used not only to improve
performance but also the reliability of response from a
safety perspective
• Prompt engineering can help identify risky behaviours of LLMs
which can help to reduce harmful behaviours and risks that may
arise from language models
• There is also a part of the community performing prompt
injection to understand the vulnerability of LLMs

Prompt Injections
• It turns out that building with LLMs, like any other systems,
comes with challenges that includes safety considerations.
• Prompt injections aim to find vulnerabilities in LLMs
• Some common issues include:
• Prompt Injection
• Prompt Leaking
• Jailbreaking

Prompt Injection
• Prompt injection is used to hijack an LM’s output by
injecting an untrusted command that overrides instruction
of a prompt
• This could easily happen if you just concatenate your
prompt with another user generated prompt

Prompt Leaking
• Prompt leaking aims to force the model to spit out
information about its own prompt.
• This can lead to leaking of either sensitive, private or
information that’s confidential

Jailbreaking
• Jailbreaking is another form of prompt injection where the
goal is to bypass safety and moderation features
• LLMs provided via APIs might be coupled with safety
features or content moderation which can be bypassed
with harmful prompts/attacks
• This might sound like a difficult task but it’s not because
the model is usually served static and might have these
vulnerabilities due to many factors such as the data it was
trained on, etc.

Jailbreaking examples
RLHF
• Reinforcement Learning from Human Feedback (RLHF) is
now being used to train LLMs that fit human preference
• RLHF involves collecting high-quality prompt datasets
• Popular examples:
• Claude (Anthropic)
• ChatGPT (OpenAI)

Source: ChatGPT: Optimizing Language Models for Dialogue

Future Directions
• Augmented LMs
• Emergent ability of LMs
• Acting / Planning - Reinforcement Learning
• Multimodal Prompting
• Graph Prompting

What’s next?
• This is the first lecture in a series of lectures
• Subscribe to the channel
• Follow us on Twitter (@dair_ai and @omarsar0) for
announcements on upcoming special lectures covering a
variety of topics in language models

• The lectures, guides, notebooks, and other prompt

engineering content will live on the Prompt Engineering
Guide:
• https://github.com/dair-ai/Prompt-Engineering-Guide

Prompt Engineering Guide

https://github.com/dair-ai/Prompt-Engineering-Guide