Positional encoding in PyTorch - Model Pipeline Trace

Model Pipeline - Positional encoding

This pipeline shows how positional encoding adds position information to input data so a model can understand order, especially in sequences like sentences.

Data Flow - 3 Stages

1Input Embeddings

10 tokens x 512 dimensions→Convert tokens to vector embeddings→10 tokens x 512 dimensions

[[0.1, 0.3, ..., 0.2], [0.05, 0.4, ..., 0.1], ..., [0.2, 0.1, ..., 0.3]]

↓

2Generate Positional Encoding

10 positions x 512 dimensions→Calculate sine and cosine values for each position and dimension→10 positions x 512 dimensions

[[sin(0/10000^(0/512)), cos(0/10000^(1/512)), ..., sin(0/10000^(510/512)), cos(0/10000^(511/512))], ..., [sin(9/10000^(0/512)), cos(9/10000^(1/512)), ..., sin(9/10000^(510/512)), cos(9/10000^(511/512))]]

↓

3Add Positional Encoding to Embeddings

10 tokens x 512 dimensions→Element-wise addition of embeddings and positional encoding→10 tokens x 512 dimensions

Embedding vector + positional encoding vector for each token

Training Trace - Epoch by Epoch

Loss
1.2 |****
1.0 |*** 
0.8 |**  
0.6 |*   
0.4 |    
     1  2  3  4  5  Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	1.2	0.45	Model starts learning with random weights and positional info
2	0.9	0.60	Loss decreases as model uses positional encoding to understand order
3	0.7	0.72	Model improves predictions by combining token meaning and position
4	0.55	0.80	Clear improvement showing positional encoding helps sequence tasks
5	0.45	0.85	Training converges with positional info aiding context understanding

Prediction Trace - 3 Layers

Layer 1: Input Embedding

Layer 2: Positional Encoding Calculation

Layer 3: Add Positional Encoding

Model Quiz - 3 Questions

Why do we add positional encoding to token embeddings?

A To increase the size of the embeddings randomly

B To remove noise from the input data

C To give the model information about the order of tokens

D To convert tokens into numbers

Key Insight

Positional encoding helps models understand the order of tokens in sequences by adding unique position information to embeddings, which improves learning and prediction accuracy in sequence tasks.

Practice

1. What is the main purpose of positional encoding in transformer models?

easy

A. To increase the number of layers in the model

B. To reduce the size of the input data

C. To add information about the order of tokens in a sequence

D. To normalize the input embeddings

Solution

1. Step 1: Understand the role of positional encoding

   Positional encoding adds information about the position of each token in the input sequence, which helps the model understand order. The other options relate to unrelated tasks such as increasing the number of layers, reducing input size, or normalizing embeddings.

2. Final Answer:

   To add information about the order of tokens in a sequence -> Option C

3. Quick Check:

   Positional encoding = order info [OK]

Hint: Positional encoding = position info added to embeddings [OK]

Common Mistakes:

• Thinking positional encoding reduces data size
• Confusing positional encoding with normalization
• Assuming it increases model layers

2. Which of the following is the correct way to compute the positional encoding for position pos and dimension i using sine and cosine functions in PyTorch?

easy

A. pos_enc[pos, i] = sin(pos * 10000^(i/d_model))

B. pos_enc[pos, 2*i] = sin(pos / 10000^(2*i/d_model)); pos_enc[pos, 2*i+1] = cos(pos / 10000^(2*i/d_model))

C. pos_enc[pos, i] = cos(pos / 10000^(i/d_model))

D. pos_enc[pos, 2*i] = cos(pos / 10000^(2*i/d_model)); pos_enc[pos, 2*i+1] = sin(pos / 10000^(2*i/d_model))

Solution

1. Step 1: Recall the positional encoding formula

   Positional encoding uses sine for even indices and cosine for odd indices with the formula: pos_enc[pos, 2*i] = sin(pos / 10000^(2*i/d_model)); pos_enc[pos, 2*i+1] = cos(pos / 10000^(2*i/d_model)).

2. Final Answer:

   pos_enc[pos, 2*i] = sin(pos / 10000^(2*i/d_model)); pos_enc[pos, 2*i+1] = cos(pos / 10000^(2*i/d_model)) -> Option B

3. Quick Check:

   Sine even, cosine odd = pos_enc[pos, 2*i] = sin(pos / 10000^(2*i/d_model)); pos_enc[pos, 2*i+1] = cos(pos / 10000^(2*i/d_model)) [OK]

Hint: Sine for even dims, cosine for odd dims in positional encoding [OK]

Common Mistakes:

• Swapping sine and cosine roles
• Using multiplication instead of division in exponent
• Applying sine or cosine to all dimensions only

3. Given the following PyTorch code snippet for positional encoding, what is the shape of the output tensor pos_encoding?

import torch
import math

def get_positional_encoding(seq_len, d_model):
    pos_encoding = torch.zeros(seq_len, d_model)
    for pos in range(seq_len):
        for i in range(0, d_model, 2):
            pos_encoding[pos, i] = math.sin(pos / (10000 ** (i / d_model)))
            if i + 1 < d_model:
                pos_encoding[pos, i + 1] = math.cos(pos / (10000 ** (i / d_model)))
    return pos_encoding

pos_encoding = get_positional_encoding(50, 512)

medium

A. [50, 512]

B. [512, 50]

C. [50, 256]

D. [512, 512]

Solution

1. Step 1: Analyze function input parameters and output shape

   The function takes seq_len=50 and d_model=512, creating a tensor of zeros with shape (50, 512). The loops fill values in this tensor without changing its shape, so output shape remains (50, 512).

2. Final Answer:

   [50, 512] -> Option A

3. Quick Check:

   Output shape = (seq_len, d_model) = [50, 512] [OK]

Hint: Output shape matches (sequence length, embedding dim) [OK]

Common Mistakes:

• Confusing dimensions order
• Thinking output is half the embedding size
• Assuming square matrix output

4. Identify the error in this PyTorch positional encoding code and select the fix:

import torch
import math

def positional_encoding(seq_len, d_model):
    pos_enc = torch.zeros(seq_len, d_model)
    for pos in range(seq_len):
        for i in range(d_model):
            pos_enc[pos, i] = math.sin(pos / 10000 ** (i / d_model))
    return pos_enc

pos_enc = positional_encoding(10, 16)

medium

A. Use sine for even indices and cosine for odd indices instead of sine only

B. Change math.sin to torch.sin for GPU compatibility

C. Initialize pos_enc with ones instead of zeros

D. Remove the inner loop and use vectorized operations

Solution

1. Step 1: Review positional encoding formula and identify fix

   Positional encoding uses sine for even dimensions and cosine for odd dimensions, but code uses sine only. Use sine for even indices and cosine for odd indices instead of sine only correctly fixes this.

2. Final Answer:

   Use sine for even indices and cosine for odd indices instead of sine only -> Option A

3. Quick Check:

   Sine and cosine alternate = Use sine for even indices and cosine for odd indices instead of sine only [OK]

Hint: Positional encoding alternates sine and cosine per dimension [OK]

Common Mistakes:

• Using sine only for all dimensions
• Ignoring cosine for odd indices
• Thinking initialization value causes error

5. You want to add positional encoding to a batch of input embeddings with shape (batch_size, seq_len, d_model) in PyTorch. Which approach correctly applies positional encoding to each sequence in the batch?

hard

A. Create a positional encoding tensor of shape (batch_size, d_model) and multiply element-wise with input embeddings

B. Create a positional encoding tensor of shape (seq_len,) and add it to input embeddings directly

C. Create a positional encoding tensor of shape (batch_size, seq_len, 1) and concatenate it to input embeddings

D. Create a positional encoding tensor of shape (seq_len, d_model) and add it to input embeddings using broadcasting

Solution

1. Step 1: Understand input and positional encoding shapes

   Input embeddings have shape (batch_size, seq_len, d_model). Positional encoding is per position and dimension, so shape (seq_len, d_model) fits. Adding it broadcasts over batch dimension correctly.

2. Final Answer:

   Create a positional encoding tensor of shape (seq_len, d_model) and add it to input embeddings using broadcasting -> Option D

3. Quick Check:

   Broadcast positional encoding over batch = Create a positional encoding tensor of shape (seq_len, d_model) and add it to input embeddings using broadcasting [OK]

Hint: Broadcast positional encoding (seq_len, d_model) over batch [OK]

Common Mistakes:

• Using positional encoding without seq_len dimension
• Multiplying instead of adding positional encoding
• Concatenating positional encoding incorrectly