Self-attention and multi-head attention in NLP - Model Pipeline Trace

Model Pipeline - Self-attention and multi-head attention

This pipeline shows how self-attention and multi-head attention help a model understand relationships between words in a sentence. It transforms input words into meaningful features, learns patterns during training, and then predicts context-aware word representations.

Data Flow - 5 Stages

1Input tokens

1 sentence x 6 words→Convert words to token IDs→1 sentence x 6 tokens

["The", "cat", "sat", "on", "the", "mat"] -> [101, 4937, 1037, 2006, 101, 3899]

↓

2Embedding lookup

1 sentence x 6 tokens→Map tokens to vectors→1 sentence x 6 tokens x 64 features

[101, 4937, 1037, 2006, 101, 3899] -> [[0.1,0.3,...], [0.2,0.1,...], ...]

↓

3Self-attention calculation

1 sentence x 6 tokens x 64 features→Compute attention scores between all tokens→1 sentence x 6 tokens x 64 features

Each token attends to all tokens, e.g., 'cat' attends to 'sat' and 'mat'

↓

4Multi-head attention

1 sentence x 6 tokens x 64 features→Split features into 4 heads, apply self-attention separately, then combine→1 sentence x 6 tokens x 64 features

4 heads each with 16 features, combined back to 64 features

↓

5Output representation

1 sentence x 6 tokens x 64 features→Final context-aware token vectors→1 sentence x 6 tokens x 64 features

Each token vector now includes info from related words

Training Trace - Epoch by Epoch

Loss
1.2 |*       
1.0 | **     
0.8 |  ***   
0.6 |   **** 
0.4 |    *****
     --------
     Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	1.2	0.45	Model starts learning basic word relationships
2	0.9	0.60	Attention weights improve, capturing more context
3	0.7	0.72	Multi-head attention helps model focus on different word aspects
4	0.5	0.80	Loss decreases steadily, accuracy rises as context understanding improves
5	0.4	0.85	Model converges with strong attention patterns

Prediction Trace - 5 Layers

Layer 1: Embedding lookup

Layer 2: Self-attention scores

Layer 3: Multi-head attention

Layer 4: Weighted sum of values

Layer 5: Final output

Model Quiz - 3 Questions

What does self-attention help the model do?

A Translate words into another language

B Understand relationships between all words in a sentence

C Reduce the number of words in a sentence

D Generate random word sequences

Key Insight

Self-attention allows the model to weigh the importance of each word relative to others, capturing context effectively. Multi-head attention enhances this by learning multiple perspectives at once, improving understanding and prediction quality.

Practice

1. What is the main purpose of self-attention in natural language processing?

easy

A. To reduce the size of the input data by removing words

B. To generate random sentences without context

C. To translate text from one language to another

D. To let the model focus on important words by comparing all words to each other

Solution

1. Step 1: Understand self-attention's role

   Self-attention helps the model look at all words in a sentence and decide which ones are important by comparing them to each other.

2. Step 2: Match purpose with options

   To let the model focus on important words by comparing all words to each other correctly describes this focus mechanism, while others describe unrelated tasks.

3. Final Answer:

   To let the model focus on important words by comparing all words to each other -> Option D

4. Quick Check:

   Self-attention = focus on important words [OK]

Hint: Self-attention means comparing words to find importance [OK]

Common Mistakes:

• Confusing self-attention with translation
• Thinking self-attention removes words
• Assuming it generates random text

2. Which of the following is the correct way to describe multi-head attention?

easy

A. Running several self-attention processes in parallel to get richer understanding

B. Applying self-attention only once on the input

C. Using attention only on the first word of a sentence

D. Ignoring word relationships and focusing on word order only

Solution

1. Step 1: Recall multi-head attention definition

   Multi-head attention means running multiple self-attention operations at the same time to capture different aspects of word relationships.

2. Step 2: Compare options to definition

   Running several self-attention processes in parallel to get richer understanding matches this exactly; others describe incomplete or incorrect ideas.

3. Final Answer:

   Running several self-attention processes in parallel to get richer understanding -> Option A

4. Quick Check:

   Multi-head attention = multiple self-attentions [OK]

Hint: Multi-head means many self-attentions at once [OK]

Common Mistakes:

• Thinking multi-head means single attention
• Believing it focuses only on first word
• Ignoring word relationships

3. Given the following simplified self-attention scores matrix for a 3-word sentence:
Scores = [[1, 0.5, 0], [0.5, 1, 0.2], [0, 0.2, 1]]
What is the attention weight for the second word attending to the third word after applying softmax on its row?

medium

A. Approximately 0.21

B. Approximately 0.50

C. Approximately 0.29

D. Approximately 0.70

Solution

1. Step 1: Extract the second row scores

   The second word's scores are [0.5, 1, 0.2].

2. Step 2: Apply softmax to these scores

   Softmax formula: exp(score) / sum(exp(all scores)). Calculate exp(0.5)=1.65, exp(1)=2.72, exp(0.2)=1.22. Sum = 1.65+2.72+1.22=5.59. Attention weight for third word = 1.22/5.59 ≈ 0.218.

3. Final Answer:

   Approximately 0.21 -> Option A

4. Quick Check:

   Softmax normalizes scores to probabilities [OK]

Hint: Softmax turns scores into probabilities summing to 1 [OK]

Common Mistakes:

• Forgetting to exponentiate scores
• Dividing by wrong sum
• Mixing row and column values

4. Consider this Python code snippet for multi-head attention weights calculation:

import numpy as np

def multi_head_attention(scores_list):
    heads = []
    for scores in scores_list:
        weights = np.exp(scores) / np.sum(np.exp(scores))
        heads.append(weights)
    return np.mean(heads, axis=0)

scores_list = [np.array([1, 0, 2]), np.array([0, 1, 1])]
print(multi_head_attention(scores_list))

What is the main bug in this code?

medium

A. Softmax is applied incorrectly; denominator should sum over exp(scores) per head

B. The function returns mean of weights instead of concatenating heads

C. The code uses np.exp twice causing overflow

D. Scores_list should be a 2D array, not a list of arrays

Solution

1. Step 1: Analyze softmax calculation

   Softmax is correctly applied per head by dividing exp(scores) by sum of exp(scores).

2. Step 2: Check output aggregation

   The function averages the weights from each head, but multi-head attention should concatenate or combine heads differently, not average weights element-wise.

3. Final Answer:

   The function returns mean of weights instead of concatenating heads -> Option B

4. Quick Check:

   Multi-head attention combines heads, not averages weights [OK]

Hint: Multi-head attention concatenates heads, not averages weights [OK]

Common Mistakes:

• Thinking averaging weights is correct
• Confusing softmax denominator
• Assuming input format is wrong

5. You want to improve a Transformer model's ability to understand complex sentences by increasing the number of attention heads from 4 to 8. What is the most likely effect of this change?

hard

A. The model will ignore word order completely

B. The model will run faster but lose accuracy

C. The model can capture more diverse word relationships but may require more computation

D. The model will only focus on the first half of the sentence

Solution

1. Step 1: Understand effect of increasing attention heads

   More heads mean the model can look at different parts of the sentence simultaneously, capturing richer relationships.

2. Step 2: Consider computational cost and accuracy

   Increasing heads usually increases computation and memory needs but can improve understanding and accuracy.

3. Final Answer:

   The model can capture more diverse word relationships but may require more computation -> Option C

4. Quick Check:

   More heads = richer focus + more compute [OK]

Hint: More heads = better focus but slower model [OK]

Common Mistakes:

• Assuming more heads always make model faster
• Thinking word order is ignored
• Believing model focuses only on part of sentence