← Back to categories

LLMs & Generative AI

Transformer Architecture

8 questions

Multiple Choice

Causal Masking Purpose

Positional Encoding Necessity

Attention Score Computation

Multi-Head Attention Benefits

Open-Ended

Self-Attention Mechanism

Positional Encoding

Multi-Head Attention

Encoder-Only vs Decoder-Only Architectures

Training & Fine-tuning

8 questions

Multiple Choice

LoRA Mechanism

RLHF Components

Catastrophic Forgetting

DPO vs RLHF

Open-Ended

Fine-tuning vs Pretraining

Pretraining Objectives

RLHF and Preference Learning

LoRA and Parameter-Efficient Fine-tuning

Prompting Techniques

8 questions

Multiple Choice

Chain-of-Thought Effectiveness

Few-Shot vs Zero-Shot

System Prompt Security

ReAct Prompting Pattern

Open-Ended

Zero-Shot vs Few-Shot Prompting

Chain-of-Thought Prompting

Prompt Engineering Best Practices

System Prompts and Roles

RAG & Knowledge Retrieval

8 questions

Multiple Choice

Embedding Model Selection

Chunking Strategy Impact

RAG vs Fine-tuning

Hybrid Search Benefits

Open-Ended

RAG Overview

Embeddings for Retrieval

Document Chunking Strategies

Vector Databases

Evaluation & Safety

8 questions

Multiple Choice

Red Teaming Purpose

Hallucination Detection

LLM Benchmark Limitations

LLM-as-Judge Considerations

Open-Ended

Understanding Hallucinations

LLM Evaluation and Benchmarks

AI Alignment

Red Teaming LLMs

Inference & Deployment

8 questions

Multiple Choice

Quantization Tradeoffs

KV Cache Purpose

Batching Strategies

Speculative Decoding

Open-Ended

Model Quantization

Knowledge Distillation

LLM Serving Strategies

Latency Optimization