MauryaHub PYQ Practice

Large Language Models · End Term · May 2025 · AN

Large Language Models · End Term · May 2025 AN

← Course papers · Start practice / exam

Questions and published explanations below are available without starting a test. Some questions may not have a published solution yet.

0 / 17 published solutions

Question 192 MCQ · 3.0 marks

Source diagram or notation
A.Source diagram or notation
B.Source diagram or notation
C.Source diagram or notation
D.Source diagram or notation

A published solution is not available for this question yet.

Report a correction for Q192

Question 193 MCQ · 2.0 marks

A GPT model is trained using **causal language modeling**. During training, for a sequence of T = 4 tokens, which of the following correctly represents the **attention mask matrix** applied to the attention logits?
A.Source diagram or notation
B.Source diagram or notation
C.Source diagram or notation
D.Source diagram or notation

A published solution is not available for this question yet.

Report a correction for Q193

Question 194 MCQ · 2.0 marks

Source diagram or notation
A.
(a) : Random Local Attention, (b) : Strided Local Attention, (c) : Sparse Block Attention, (d) : Local + Global Attention
B.
(a) : Strided Local Attention, (b) : Local + Global Attention, (c) : Random Local Attention, (d) : Sparse Block Attention
C.
(a) : Strided Local Attention, (b) : Random Local Attention, (c) : Local + Global Attention, (d) : Sparse Block Attention
D.
(a) : Sparse Block Attention, (b) : Local + Global Attention, (c) : Strided Local Attention, (d) : Random Local Attention

A published solution is not available for this question yet.

Report a correction for Q194

Question 195 MSQ · 3.0 marks

Which of the following preprocessing steps are commonly used when preparing data for transformer models like BERT?
A.
Adding special tokens such as [CLS] and [SEP] to the sequence
B.
Splitting tokens into subword units using a model-specific tokenizer
C.
Removing all punctuation marks to ensure cleaner embeddings
D.
Adding positional information to each token embedding

A published solution is not available for this question yet.

Report a correction for Q195

Question 196 MSQ · 3.0 marks

Which modifications are used in transformers to handle longer sequences efficiently?
A.
Using sparse attention patterns to reduce computational complexity
B.
Applying low-rank matrix factorization to approximate attention
C.
Increasing only the feed-forward network width without modifying attention
D.
Replacing the softmax function in attention with other activation function for faster training

A published solution is not available for this question yet.

Report a correction for Q196

Question 197 MSQ · 3.0 marks

In Transformer models that use relative position embeddings (such as in Transformer-XL or T5), clipping is often applied to the relative position indices. Which of the following statements about clipping in relative position embeddings are correct?
A.
Clipping ensures that very large relative distances are mapped to a fixed maximum distance.
B.
Without clipping, the model would require embeddings for every possible relative distance, which is infeasible for long sequences.
C.
Clipping increases the precision of embeddings for large relative distances.
D.
Clipping introduces an upper bound k such that all distances greater than k are mapped to the same index.

A published solution is not available for this question yet.

Report a correction for Q197

Question 198 NAT · 2.0 marks

Source diagram or notation

A published solution is not available for this question yet.

Report a correction for Q198

Question 199 NAT · 3.0 marks

Source diagram or notation
For the input “you enjoy tea often”, compute the final representation of the word “enjoy” after the attention layer (i.e., after applying WQ, WK, WV , attention weights, and WO). Enter the sum of the elements in the resulting vector.

A published solution is not available for this question yet.

Report a correction for Q199

Question 200 NAT · 2.0 marks

Source diagram or notation
Suppose the input sentence is “tea you enjoy often”. Using the same matrices and processing method, what is the attention score (i.e., softmax entry from A) for the query word “tea” attending to key word “often”?

A published solution is not available for this question yet.

Report a correction for Q200

Question 201 MCQ · 2.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
Choose the correct representation of π for the given mask image.
A.
π = (2, 1, 3, 4, 0)
B.
π = (0, 4, 3, 1, 2)
C.
π = (0, 3, 4, 2, 1)
D.
All of these

A published solution is not available for this question yet.

Report a correction for Q201

Question 202 NAT · 1.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
How many permutations of π are possible for n = 5?

A published solution is not available for this question yet.

Report a correction for Q202

Question 203 NAT · 2.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
What percentage of entries in the attention matrix are zero (i.e., the sparsity)?

A published solution is not available for this question yet.

Report a correction for Q203

Question 204 MCQ · 2.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
Source diagram or notation
A.
(0, 1, 2, 3, 4, 5)
B.
(-1, 0, 1, 2, 3, 4)
C.
(-3, -2, -1, 0, 1, 2)
D.
(-2, -1, 0, 1, 2, 3)

A published solution is not available for this question yet.

Report a correction for Q204

Question 205 MCQ · 2.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
Source diagram or notation
A.
[−0.4,−0.4,−0.4,−0.4]
B.
[−0.1,−0.1,−0.1,−0.1]
C.
[−0.8,−0.8,−0.8,−0.8]
D.
[−0.2,−0.2,−0.2,−0.2]

A published solution is not available for this question yet.

Report a correction for Q205

Question 206 NAT · 3.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
Source diagram or notation

A published solution is not available for this question yet.

Report a correction for Q206

Question 207 NAT · 2.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
Source diagram or notation

A published solution is not available for this question yet.

Report a correction for Q207

Question 208 NAT · 3.0 marks

Source diagram or notation
Based on the above data, answer the given subquestions.
Source diagram or notation

A published solution is not available for this question yet.

Report a correction for Q208