LoRA and QLoRA concepts in Prompt Engineering / GenAI - Cheat Sheet & Quick Revision

Recall & Review

beginner

What does LoRA stand for in machine learning?

LoRA stands for Low-Rank Adaptation. It is a method to efficiently fine-tune large models by updating only small, low-rank matrices instead of the full model.

intermediate

How does LoRA reduce the number of parameters to update during fine-tuning?

LoRA inserts small low-rank matrices into the model layers and only trains these matrices. This reduces the number of parameters updated compared to training the entire model.

intermediate

What is QLoRA and how does it differ from LoRA?

QLoRA stands for Quantized LoRA. It combines LoRA's low-rank adaptation with quantization, which reduces model size by using fewer bits per parameter, enabling fine-tuning on smaller hardware.

beginner

Why is quantization useful in QLoRA?

Quantization reduces the memory and compute needed by representing model weights with fewer bits. This makes it possible to fine-tune large models on less powerful devices.

beginner

In simple terms, how would you explain the benefit of using LoRA or QLoRA?

They let you update a big model quickly and with less computer power by only changing small parts of it, making it easier and cheaper to customize AI models.

What is the main goal of LoRA in model fine-tuning?

A To update only small low-rank matrices instead of the full model

B To increase the size of the model

C To remove layers from the model

D To train the entire model from scratch

LoRA focuses on updating small low-rank matrices to reduce the number of parameters trained.

What does QLoRA add to the LoRA method?

A Quantization to reduce model size and memory use

B More layers to the model

C A new optimizer

D Data augmentation techniques

QLoRA combines LoRA with quantization to make fine-tuning more memory efficient.

Why is quantization helpful for fine-tuning large models?

A It makes the model bigger

B It increases model accuracy automatically

C It removes the need for training data

D It reduces memory and compute requirements

Quantization reduces the size of model weights, lowering memory and compute needs.

Which of these is NOT a benefit of LoRA?

A Faster fine-tuning with fewer parameters

B Trains the entire model from scratch

C Easier to customize large models

D Requires less computer memory

LoRA does not train the entire model from scratch; it updates small parts only.

LoRA is best described as a method to:

A Generate new training data

B Compress data before training

C Adapt large models efficiently by training small matrices

D Replace neural networks with decision trees

LoRA adapts large models by training small low-rank matrices efficiently.

Explain in your own words what LoRA is and why it helps with fine-tuning large AI models.

Think about how updating fewer parts of a big model can save time and memory.

Describe how QLoRA improves on LoRA and what problem it solves.

Focus on how using fewer bits per parameter helps with hardware limits.

Practice

1. What is the main purpose of LoRA in training large AI models?

easy

A. To increase the size of the model for better accuracy

B. To add small trainable parts that make training easier and cheaper

C. To replace the entire model with a smaller one

D. To remove layers from the model to speed up training

Solution

1. Step 1: Understand LoRA's role in model training

   LoRA adds small trainable parts to a big model instead of retraining the whole model, making training easier and cheaper.

2. Step 2: Compare options with LoRA's purpose

   Options B, C, and D describe changing model size or structure, which is not what LoRA does.

3. Final Answer:

   To add small trainable parts that make training easier and cheaper -> Option B

4. Quick Check:

   LoRA = small trainable parts for easier training [OK]

Hint: LoRA adds small parts to big models for easier training [OK]

Common Mistakes:

• Thinking LoRA replaces the whole model
• Confusing LoRA with model size increase
• Assuming LoRA removes layers

2. Which of the following correctly describes QLoRA?

easy

A. A method that combines LoRA with quantization to save memory

B. A technique that trains models without any compression

C. A way to increase model size by adding layers

D. A method that removes LoRA parts to speed up training

Solution

1. Step 1: Recall QLoRA's definition

   QLoRA combines LoRA with quantization (number compression) to reduce memory use and speed up training.

2. Step 2: Eliminate incorrect options

   Options B, C, and D contradict QLoRA's purpose by ignoring compression or removing LoRA parts.

3. Final Answer:

   A method that combines LoRA with quantization to save memory -> Option A

4. Quick Check:

   QLoRA = LoRA + quantization for memory saving [OK]

Hint: QLoRA = LoRA plus compression to save memory [OK]

Common Mistakes:

• Ignoring quantization in QLoRA
• Thinking QLoRA removes LoRA parts
• Believing QLoRA increases model size

3. Given this Python snippet using LoRA and QLoRA concepts:

model_size = 1000  # in MB
lora_size = 10    # LoRA adds 10 MB
quantization_factor = 0.25  # QLoRA compresses to 25%

lora_model_size = model_size + lora_size
qlora_model_size = int(lora_model_size * quantization_factor)
print(qlora_model_size)

What is the printed output?

medium

A. 252

B. 250

C. 260

D. 275

Solution

1. Step 1: Calculate LoRA model size

   LoRA adds 10 MB to 1000 MB, so lora_model_size = 1000 + 10 = 1010 MB.

2. Step 2: Apply QLoRA compression

   QLoRA compresses to 25%, so qlora_model_size = int(1010 * 0.25) = int(252.5) = 252 MB.

3. Final Answer:

   252 -> Option A

4. Quick Check:

   1010 * 0.25 = 252.5 -> 252 [OK]

Hint: Add LoRA size, then multiply by compression factor [OK]

Common Mistakes:

• Multiplying before adding LoRA size
• Rounding incorrectly
• Using 0.2 instead of 0.25 for compression

4. This code tries to calculate QLoRA model size but has an error:

model_size = 800
lora_size = 20
quantization_factor = 0.3

qlora_model_size = model_size + lora_size * quantization_factor
print(qlora_model_size)

What is the error and how to fix it?

medium

A. Wrong variable name; change quantization_factor to quant_factor

B. No error; code is correct

C. Should use integer division // instead of *

D. Missing parentheses; fix with (model_size + lora_size) * quantization_factor

Solution

1. Step 1: Identify operator precedence issue

   Multiplication (*) happens before addition (+), so only lora_size is multiplied by quantization_factor, not the sum.

2. Step 2: Fix with parentheses

   Use (model_size + lora_size) * quantization_factor to multiply the total size by compression factor.

3. Final Answer:

   Missing parentheses; fix with (model_size + lora_size) * quantization_factor -> Option D

4. Quick Check:

   Parentheses fix operator order [OK]

Hint: Use parentheses to control addition before multiplication [OK]

Common Mistakes:

• Ignoring operator precedence
• Changing variable names incorrectly
• Using wrong operators like //

5. You want to fine-tune a large language model on a small laptop with limited memory. Which approach best balances training speed and memory use?

hard

A. Only add LoRA layers without any compression

B. Train the full large model from scratch without compression

C. Use QLoRA to compress the model and add LoRA layers for efficient training

D. Use full precision training without LoRA or compression

Solution

1. Step 1: Understand resource limits

   Small laptops have limited memory, so full model training or full precision is too heavy.

2. Step 2: Choose best method

   QLoRA combines LoRA's small trainable parts with quantization compression, saving memory and speeding training.

3. Step 3: Compare options

   Options B and D ignore memory limits; A lacks compression benefits.

4. Final Answer:

   Use QLoRA to compress the model and add LoRA layers for efficient training -> Option C

5. Quick Check:

   QLoRA = LoRA + compression for small devices [OK]

Hint: Combine LoRA and compression for small device training [OK]

Common Mistakes:

• Ignoring compression benefits
• Trying full model training on small memory
• Using only LoRA without compression