da5013_2025T1_Q1_NA.pdf

Deep Learning Practice · Quiz 1 · Jan 2025

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

Question 213 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_2_0]]

["hugging", "face", "is", "awesome"]
[102, 463, 509, 101, 2020]
["hugging", "face", "is", "awesome", ""]
"Hugging Face is awesome!"

A published solution is not available for this question yet.

Question 214 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_3_1]]

The tokenizer will output the token [UNK] because "nonexistentword" is not in the vocabulary.
The tokenizer will output the word "nonexistentword" as a single token.
The tokenizer will output "non", "existent", "word" as separate tokens since it has split the word into known subwords.
The tokenizer will output "non", "exist", "ent", "word" as separate tokens since it has split the word into known subwords.
The tokenizer will output an error because "nonexistentword" is not present in the vocabulary and the unk_token was not defined.

A published solution is not available for this question yet.

Question 215 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_4_2]]

[2, 3]
[3, 0, 2]
[2, 0, 1, 3]
[3, 1, 0, 2]

A published solution is not available for this question yet.

Question 216 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_5_3]]

[[IMAGE:e84292cfb475ac7d_5_4]]
[[IMAGE:e84292cfb475ac7d_5_5]]
[[IMAGE:e84292cfb475ac7d_5_6]]
[[IMAGE:e84292cfb475ac7d_5_7]]

A published solution is not available for this question yet.

Question 217 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_6_8]]

[[IMAGE:e84292cfb475ac7d_6_9]]
[[IMAGE:e84292cfb475ac7d_6_10]]
[[IMAGE:e84292cfb475ac7d_6_11]]
[[IMAGE:e84292cfb475ac7d_6_12]]

A published solution is not available for this question yet.

Question 218 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_6_13]]

LoRA will be applied to the entire model, including the embedding layers.
LoRA will only be applied to the final classification layer of the model.
LoRA will not be applied to any layer of the model.
The model will throw an error because an empty list is invalid.

A published solution is not available for this question yet.

Question 219 MCQ · 4.0 marks

[[IMAGE:e84292cfb475ac7d_8_14]]

**n_t_5k > n_t_50k** because with a smaller vocabulary size, the tokenizer will split more words into smaller subwords, leading to a higher number of tokens.
**n_t_5k < n_t_50k** because a smaller vocabulary size means the tokenizer can only split the words into fewer subwords, resulting in fewer tokens.
**n_t_5k = n_t_50k** because the number of tokens is independent of the vocabulary size and depends on the input words.
**n_t_5k = n_t_50k** because both vocabularies will represent the word "unhappiness" with the same number of tokens due to the use of subword tokenization.

A published solution is not available for this question yet.

Question 220 MCQ · 2.0 marks

What type of language modeling objective is used during the pretraining of GPT-2?

Masked Language Modeling (MLM)
Sequence-to-Sequence Modeling
Causal Language Modeling (CLM)
Text Classification Objective

A published solution is not available for this question yet.

Question 221 MSQ · 3.0 marks

Which of the following is/are gradient based (fine tuning) methods?

Zero-shot Tuning
Few-shots Tuning
Instruction Fine tuning
Parameter Efficient Fine Tuning
Preference Tuning Via RLHF

A published solution is not available for this question yet.

Question 222 MSQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_9_15]]

[[IMAGE:e84292cfb475ac7d_9_16]]
[[IMAGE:e84292cfb475ac7d_9_17]]
[[IMAGE:e84292cfb475ac7d_10_18]]
[[IMAGE:e84292cfb475ac7d_10_19]]

A published solution is not available for this question yet.

Question 223 MSQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_10_20]] Select all the correct statements.

[[IMAGE:e84292cfb475ac7d_10_21]]
[[IMAGE:e84292cfb475ac7d_10_22]]
[[IMAGE:e84292cfb475ac7d_10_23]]
[[IMAGE:e84292cfb475ac7d_10_24]]

A published solution is not available for this question yet.

Question 224 MSQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_11_25]] Select all the correct statements.

[[IMAGE:e84292cfb475ac7d_11_26]]
[[IMAGE:e84292cfb475ac7d_11_27]]
[[IMAGE:e84292cfb475ac7d_11_28]]
[[IMAGE:e84292cfb475ac7d_11_29]]
[[IMAGE:e84292cfb475ac7d_11_30]]

A published solution is not available for this question yet.

Question 225 NAT · 3.0 marks

[[IMAGE:e84292cfb475ac7d_11_31]] The original train split contains 25,000 samples, and 60% of the samples have a text length greater than 200. If the dataset's text length distribution is uniform across all splits, how many samples will the subset dataset contain?

A published solution is not available for this question yet.

Question 226 NAT · 4.0 marks

[[IMAGE:e84292cfb475ac7d_13_32]] Based on the above data, answer the given subquestions.

Calculate the number of parameters in the query, key, and value matrices of a single transformer block.

A published solution is not available for this question yet.

Question 227 NAT · 4.0 marks

[[IMAGE:e84292cfb475ac7d_13_32]] Based on the above data, answer the given subquestions.

Calculate the number of embedding parameters in the model. Enter your answer in millions, rounded to two decimal places.

A published solution is not available for this question yet.

Question 228 MCQ · 3.0 marks

[[IMAGE:e84292cfb475ac7d_13_32]] Based on the above data, answer the given subquestions.

What does the n_ctx parameter in the GPT-2 configuration represent?

The number of attention heads in the model.
The maximum length of the input sequence in tokens.
The embedding size of each token.
The total number of parameters in the model. **Managerial Economics** **Section Id :** 64065379959 **Section Number :** 12 **Section type :** Online **Mandatory or Optional :** Mandatory **Number of Questions :** 9 **Number of Questions to be attempted :** 9 **Section Marks :** 25 **Display Number Panel :** Yes

A published solution is not available for this question yet.