*LARGE LANGUAGE MODELS (LLMs)*

1. What is an LLM?

Answer:
An LLM (Large Language Model) is a type of Artificial Intelligence (AI) model trained on massive amounts of text data to understand, process, and generate human language. It uses deep learning to perform tasks such as answering questions, summarizing text, translating languages, generating code, and creating human-like conversations.

Examples:
• ChatGPT (OpenAI)
• Gemini (Google)
• Claude (Anthropic)
• Llama (Meta)
• Mistral AI

--------------------------------------------------

2. Why is it called a Large Language Model?

Answer:
• Large – Trained on enormous amounts of text data and contains billions of parameters.
• Language – Understands and generates human language.
• Model – A trained deep learning model that learns patterns from data.

--------------------------------------------------

3. What are LLMs used for?

Answer:
• Chatbots and Virtual Assistants
• Content Generation
• Code Generation
• Language Translation
• Text Summarization
• Email Writing
• Question Answering
• Sentiment Analysis
• Customer Support

--------------------------------------------------

4. Name some popular LLMs.

Answer:
• ChatGPT
• Gemini
• Claude
• Llama
• DeepSeek
• Mistral AI

--------------------------------------------------

5. How does an LLM work?

Answer:

Large Text Dataset
        ↓
Model Training
        ↓
Transformer Architecture
        ↓
Trained LLM
        ↓
User Prompt
        ↓
Generated Response

The model predicts the next most probable token based on the given context.

--------------------------------------------------

6. What is a Transformer?

Answer:
A Transformer is a deep learning architecture introduced by Google in 2017 through the paper "Attention Is All You Need." It uses a mechanism called Self-Attention to understand relationships between words in a sentence. Most modern LLMs, including ChatGPT, Gemini, Claude, and Llama, are based on the Transformer architecture.

--------------------------------------------------

7. What is Self-Attention?

Answer:
Self-Attention helps an LLM understand how words in a sentence relate to one another, allowing it to capture context more accurately.

Example:
"The animal didn't cross the street because it was tired."

Here, Self-Attention helps the model understand that "it" refers to the animal, not the street.

--------------------------------------------------

8. What are Tokens?

Answer:
Tokens are the small units of text processed by an LLM. A token can represent a whole word, part of a word, or punctuation.

Example:
"I love Python."

This sentence is split into multiple tokens before processing.

--------------------------------------------------

9. What are Parameters?

Answer:
Parameters are the learned weights inside an AI model that determine how it makes predictions. During training, these parameters are adjusted to improve the model's accuracy. Modern LLMs often contain billions of parameters.

--------------------------------------------------

10. What is Prompt Engineering?

Answer:
Prompt Engineering is the practice of writing clear, specific, and effective prompts to obtain accurate and useful responses from an LLM.

Example:

Poor Prompt:
Write Python.

Better Prompt:
Write a Python program to check whether a number is prime and explain each step.

--------------------------------------------------

11. What is a Prompt?

Answer:
A prompt is the instruction, question, or input given to an AI model to generate a response.

Example:
Explain Object-Oriented Programming in Python.

--------------------------------------------------

12. What is Fine-Tuning?

Answer:
Fine-Tuning is the process of taking a pre-trained model and training it further on a domain-specific dataset so that it performs better for a particular task.

Example:

General LLM
     ↓
Medical Dataset
     ↓
Medical Chatbot

--------------------------------------------------

13. What is RAG (Retrieval-Augmented Generation)?

Answer:
RAG combines a Large Language Model with an external knowledge source. Instead of relying only on its training data, it retrieves relevant information from documents or databases before generating a response.

Workflow:

Question
    ↓
Retrieve Relevant Documents
    ↓
LLM
    ↓
Answer

--------------------------------------------------

14. What is Hallucination?

Answer:
Hallucination occurs when an LLM generates incorrect or fabricated information while presenting it confidently as if it were true.

--------------------------------------------------

15. How can Hallucinations be reduced?

Answer:
• Write better prompts
• Use RAG
• Fine-Tuning
• Human verification
• Reliable external data sources

--------------------------------------------------

16. What is an AI Agent?

Answer:
An AI Agent is an intelligent system that can plan, reason, make decisions, and use external tools to complete tasks with minimal human intervention.

Example:

Travel Request
      ↓
Book Flights
      ↓
Book Hotels
      ↓
Generate Itinerary

--------------------------------------------------

17. Difference between AI, ML, and LLM

AI:
The broad field of creating intelligent systems.

ML:
A subset of AI that enables systems to learn patterns from data.

LLM:
A specialized AI model designed to understand and generate human language.

--------------------------------------------------

18. What is Generative AI?

Answer:
Generative AI is a branch of Artificial Intelligence that creates new content such as text, images, audio, video, or code.

Examples:
• ChatGPT
• Gemini
• DALL·E
• GitHub Copilot

--------------------------------------------------

19. LLM vs Traditional Chatbot

Traditional Chatbot:
• Rule-based
• Fixed responses
• Limited understanding

LLM:
• Learns from massive datasets
• Generates dynamic responses
• Understands context and intent

--------------------------------------------------

20. Benefits of LLMs

• Human-like conversations
• Faster content generation
• Coding assistance
• Translation
• Summarization
• Automation
• Better customer support

--------------------------------------------------

21. Limitations of LLMs

• Can hallucinate
• May contain biases
• High computational cost
• Knowledge may not always be current
• Requires significant computing resources

--------------------------------------------------

22. What did you learn during your CloudKarya GenAI Internship?

Answer:
During my CloudKarya Generative AI Internship, I learned the fundamentals of Large Language Models (LLMs), Prompt Engineering, Retrieval-Augmented Generation (RAG), AI workflows, and Responsible AI practices. I also gained practical exposure to how LLMs are integrated into real-world applications such as chatbots, automation systems, and intelligent assistants.

--------------------------------------------------

23. Difference between ChatGPT and Google Search

Answer:
Google Search retrieves relevant web pages and links based on a user's query, whereas ChatGPT generates conversational responses using its trained knowledge and, when available, retrieved information. Google helps users find information, while ChatGPT helps explain, summarize, and generate content interactively.

--------------------------------------------------

TCS INTERVIEW RAPID-FIRE

• What is an LLM?
A Large Language Model is a deep learning model trained on massive text datasets to understand and generate human language.

• Name some LLMs.
ChatGPT, Gemini, Claude, Llama, DeepSeek, Mistral AI.

• What is Prompt Engineering?
Designing effective prompts to obtain accurate and useful responses from an LLM.

• What is RAG?
Retrieval-Augmented Generation combines an LLM with external knowledge retrieval to produce more accurate and up-to-date responses.

• What is Hallucination?
When an LLM generates incorrect or fabricated information while presenting it confidently as factual.

• What is Fine-Tuning?
Further training a pre-trained model on domain-specific data to improve its performance for a specific task.

• What is a Transformer?
The deep learning architecture that powers most modern LLMs using Self-Attention to understand context.

--------------------------------------------------

A Large Language Model (LLM) is a deep learning model trained on massive amounts of text data to understand and generate human language. It is based on the Transformer architecture and is capable of tasks such as question answering, summarization, translation, code generation, and content creation. Examples include ChatGPT, Gemini, Claude, and Llama. During my CloudKarya GenAI internship, I learned about LLMs, prompt engineering, AI workflows, and how these models are applied in real-world solutions.