cs4003_2025T3_Q1_NA.pdf

You are given a GPT-style model with a vocabulary size of 50,257, a context length of 1024, and an embedding dimension of 768. The model learns its positional embeddings rather than using fixed sinusoidal ones. Calculate the total number of parameters in the model’s embedding layer, which includes both token embeddings and positional embeddings. Report your answer in millions, rounded to one decimal place.(in M)

1 Objective Type [1 Mark each] 1, As discussed in lectures, which of the following is not a type of fake news? (MCQ) A. News Satire B. News Parody C. News Distortion D. News Fabrication 2. As discussed in lectures, what is the ‘Imphed Truth Effect’ in the context of misinformation? (MCQ) A. The tendency to believe true information as false due to frequent exposure. B, The tendency to believe false information to be true after repeated exposure. C, The assumption that all information on the internet is true, D, The tendency to doubt all forms of digital content. 3. As discussed in lectures, In the Information Disorder framework, disinformation is defined as (MCQ) A. False information shared with no intent to harm. B, True information used to harm someone, C, False information shared intentionally to cause harm. D. Misleading headlines with legitimate sources. 4, Which of the following is not a type of cognitive bias as discussed in lectures? (MCQ) A. Popularity cues B. Persistence of inaccuracy C. Confirmation bias D, Acting without reading 5. What is the full form of IFCN? (MCQ) A. International Fake-news Control Network B, Internet Fact Collaboration Network C, International Fact-Checking Network D, International Forum for Communication News [[IMAGE:259c0f34733c8c18_4_1]]

[[IMAGE:259c0f34733c8c18_5_2]]