LLM

Articles in this section

• Introduction to Tokenization and Text Representation in NLP

• Understanding Word Embeddings: Word2Vec, GloVe, and Beyond

• Sequence Models Before Transformers: RNNs, LSTMs, and GRUs

• The Transformer Architecture: Attention, Multi-Head Attention, and Positional Encoding

• Transformer Blocks: Encoder and Decoder Stacks Explained

• Multi-Head Attention and Positional Encoding in Transformers

• Full Transformer Architecture and Training: From Input to Output

• BERT: Encoder-Only Pre-Training and Fine-Tuning Paradigm

• GPT: Decoder-Only Pre-Training and Generative Language Modeling

• Data, Scaling Laws, and Efficient Architectures for LLM Pre-Training

• Full Fine-tuning vs. Parameter-Efficient Fine-Tuning (PEFT) in LLMs

• Parameter-Efficient Fine-Tuning: LoRA and Adapter Modules

• Prompt Tuning, Prefix Tuning, and P-Tuning: Soft Prompts for LLM Adaptation

• Instruction Tuning: Making LLMs Follow Instructions

• RLHF: The Core Loop – SFT, Reward Modeling, and PPO

• Introduction to RAG: Why and How

• The Retriever Component in RAG: Vector Databases and Semantic Search

• The Generator Component in RAG: Prompt Engineering and Context Integration

• Building a Basic End-to-End RAG System

• Evaluating RAG Systems and Iterative Improvement

• Advanced RAG Techniques and Memory Concepts

• Current Challenges, Future of RAG, and Consolidation