Showing 221–240 of 653

Q221.Which is the smallest unit of memory/data in computer?

Easy
A Bit
B Byte
C Mega bite
D None of these
Hardware PPSC

Correct Answer: A

A bit (binary digit) representing a 0 or a 1 is the absolute smallest unit of digital memory.

Q222.Which device is commonly used to copy a file from one computer to another?

Easy
A USB Flash Drive
B Monitor
C Printer
D None of these
Hardware PPSC

Correct Answer: A

A USB Flash Drive (or pen drive) is a highly portable, non-volatile solid-state storage device used to transfer files between computers.

Q223.Which of the following is a volatile memory?

Easy
A RAM
B ROM
C Cache
D Both A and C
Hardware PPSC

Correct Answer: D

Both RAM (Random Access Memory) and Cache memory are volatile, meaning their stored contents are completely erased when power is lost.

Q224.What is the function of a Hard Disk Drive (HDD)?

Easy
A To display images and videos on a screen
B To process data and perform calculations
C To store data and programs permanently
D None of these
Hardware PPSC

Correct Answer: C

A Hard Disk Drive (HDD) is a non-volatile magnetic secondary storage device used to store operating systems, programs, and user files long-term.

Q225.In AI, a 'heuristic' is best described as

Easy
A A rule-of-thumb to guide search efficiently
B A data type
C An exact solution
D A hardware component
Hardware PPSC

Correct Answer: A

A heuristic is an approximate search technique or rule-of-thumb that directs problem-solving toward quick, optimal solutions rather than guarantees.

Q226.Which activation function outputs values between 0 and 1?

Easy
A Tanh
B Linear
C Sigmoid
D ReLU
Hardware PPSC

Correct Answer: C

The mathematical Sigmoid activation function squashes its inputs into a clean probability range between 0 and 1.

Q227.RAM is a ______ memory.

Easy
A Permanent
B Volatile
C Primary Memory
D None of these
Hardware PPSC

Correct Answer: B

RAM (Random Access Memory) is volatile primary memory, meaning it requires active electric currents to preserve its stored data.

Q228.Cache memory is located on ______?

Easy
A CPU
B Software
C Window
D None of these
Hardware PPSC

Correct Answer: A

Cache memory is a small, ultra-fast static RAM (SRAM) unit located directly inside the CPU package to store frequently used data.

Q229.Which risk occurs when an LLM reveals sensitive information from its training or prompt data?

Easy
A Data/privacy leakage
B Tokenization
C Image compression
D Overclocking
Hardware PPSC

Correct Answer: A

Data leakage occurs when an AI model exposes private training data or confidential user inputs in its responses.

Q230.What is a system prompt usually used for?

Easy
A Setting model behavior, rules, boundaries, or instructions
B Opening a file
C Installing software
D Changing monitor color
Hardware PPSC

Correct Answer: A

A system prompt is a high-level instruction used to establish the overall role, boundaries, tone, and security rules for the AI.

Q231.What is few-shot prompting?

Easy
A Providing a few examples of expected input/output in the prompt
B Giving no prompt
C Deleting examples
D Using only images
Hardware PPSC

Correct Answer: A

Few-shot prompting is a technique where a few sample tasks are demonstrated inside the prompt to teach the model how to perform a task.

Q232.What is zero-shot prompting?

Easy
A Giving no task
B Asking a model to perform a task without providing any examples
C Training with many examples
D Removing tokens
Hardware PPSC

Correct Answer: B

Zero-shot prompting relies on the model's pre-trained knowledge to perform a task directly from an instruction, with zero samples.

Q233.What is fine-tuning?

Easy
A Further training a pre-trained model on specific domain data
B Formatting a disk
C Deleting prompts
D Increasing screen size
Hardware PPSC

Correct Answer: A

Fine-tuning is the process of adjusting the weights of a pre-trained model by training it further on a smaller, domain-specific dataset.

Q234.What should you do after receiving an important factual answer from an LLM?

Easy
A Trust it blindly
B Verify it independently using authoritative sources
C Delete it immediately
D Ignore it
Hardware PPSC

Correct Answer: B

Due to the risk of hallucinations, any critical or factual claim made by an LLM must be verified against primary sources before being relied upon.

Q235.Which of the following is an example of a permanent storage device?

Easy
A RAM
B Hard disk
C DRAM
D None of these
Hardware PPSC

Correct Answer: B

Hard disks (and SSDs) are non-volatile permanent storage devices, whereas RAM and DRAM are volatile temporary memories.

Q236.A printer is an example of what type of device?

Easy
A Hard copy output device
B Soft copy output device
C Photocopy input device
D None of these
Hardware PPSC

Correct Answer: A

Printers and plotters produce permanent physical output on paper, which is classified as a hard copy.

Q237.Which of the following best describes 'reinforcement learning'?

Easy
A Learning from clustering
B Learning from data compression
C Learning from trial-and-error actions via rewards and penalties
D Learning from labeled historical data
Hardware PPSC

Correct Answer: C

Reinforcement learning involves an agent interacting with an environment, learning to make decisions by maximizing cumulative rewards.

Q238.What does 'OCR' stand for in AI?

Easy
A Object Class Ranking
B Output Control Routine
C Optical Character Recognition
D Online Code Repository
Hardware PPSC

Correct Answer: C

OCR stands for Optical Character Recognition, a technology used to convert printed or handwritten text images into editable, machine-encoded text.

Q239.Which of the following represents a supervised learning task?

Easy
A Anomaly grouping
B Dimensionality reduction
C Clustering customers
D Predicting house prices from labeled training data
Hardware PPSC

Correct Answer: D

Predicting numeric values (regression) or class labels using historical, labeled input-output pairs is a classic supervised learning task.

Q240.What is the main idea of 'few-shot learning'?

Easy
A Training with millions of samples
B Learning or performing tasks from only a handful of examples
C No training data needed at all
D Only using image datasets
Hardware PPSC

Correct Answer: B

Few-shot learning allows an AI model to recognize patterns or execute tasks after seeing only a very limited number of demonstrations.