BSDA5004 · Knowledge Base
Llms Notes
14
Concepts
0
Facts
0
Procedures
concept
1m
165 - Transformer Architecture — Complete Introduction
concept
1m
166 - Self-Attention and QKV Computation
concept
1m
167 - Multi-Head Attention Deep Dive
concept
1m
168 - Positional Encoding — Sinusoidal and Learned Embeddings
concept
1m
169 - Decoder Layer, Cross-Attention, and Teacher Forcing
concept
1m
170 - Layer Normalization & Residual Connections in Transformers
concept
1m
171 - Pre-training Objectives- Causal LM, Masked LM, and Beyond
concept
1m
172 - GPT Architecture- Decoder-Only Transformer & Autoregressive Generation
concept
1m
173 - BERT Architecture- Encoder-Only & Bidirectional Context
concept
1m
174 - Tokenization- BPE, WordPiece, SentencePiece
concept
1m
175 - Fine-tuning Methods- LoRA, Adapters, and PEFT
concept
1m
176 - Prompt Engineering- In-Context Learning, Chain-of-Thought
concept
1m
177 - RLHF, Alignment, and Constitutional AI
concept
1m