LoRA and QLoRA concepts in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to define a low-rank adaptation (LoRA) layer with a rank of 4.

Prompt Engineering / GenAI

lora_layer = LoRA(rank=[1])

Drag options to blanks, or click blank then click option'

A128

B16

C64

D4

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing a very large rank defeats the purpose of LoRA's parameter efficiency.

Using zero or negative values for rank.

LoRA uses a small rank value like 4 to reduce the number of trainable parameters while adapting large models.

2fill in blank

medium

Complete the code to quantize a model using QLoRA with 4-bit precision.

Prompt Engineering / GenAI

quantized_model = QLoRA(model, bits=[1])

Drag options to blanks, or click blank then click option'

A8

B2

C4

D16

Attempts:

3 left

💡 Hint

Common Mistakes

Using 8-bit quantization which is not QLoRA's typical setting.

Using 2-bit which is too low and uncommon for QLoRA.

QLoRA uses 4-bit quantization to reduce memory usage while maintaining model quality.

3fill in blank

hard

Fix the error in the code to apply LoRA adapters to a transformer model.

Prompt Engineering / GenAI

model = Transformer()
model.apply_adapters(adapter_type=[1])

Drag options to blanks, or click blank then click option'

A'LoRA'

B'quantization'

C'dropout'

D'batchnorm'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'quantization' which is unrelated to adapter application.

Using 'dropout' or 'batchnorm' which are regularization techniques, not adapters.

To apply LoRA adapters, the adapter_type must be set to 'LoRA'. Other options are unrelated.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that stores LoRA adapter weights only if their norm is greater than 0.1.

Prompt Engineering / GenAI

adapter_weights = {name: weight for name, weight in model.adapters.items() if weight.norm() [1] [2]

Drag options to blanks, or click blank then click option'

A>

B<

C0.1

D1.0

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' instead of '>' which selects weights with small norms.

Using 1.0 which is a higher threshold and may exclude relevant weights.

We want weights with norm greater than 0.1, so the condition is weight.norm() > 0.1.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps adapter names to their scaled weights if the scale is less than 0.5.

Prompt Engineering / GenAI

scaled_adapters = {name: weight * [1] for name, weight in model.adapters.items() if [2] [3] 0.5}

Drag options to blanks, or click blank then click option'

A0.1

Bscale

C<

Dweight

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'weight' in the condition instead of 'scale'.

Using '>' instead of '<' which reverses the condition.

We multiply weights by 0.1 and filter where scale < 0.5, so blanks are 0.1, scale, and <.

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