Llms
- 1.What an LLM Actually Is (Not Another Prompting Course)
- 2.Subword Tokenization: Byte-Pair Encoding
- 3.Embeddings & Positional Encoding at Scale
- 4.Self-Attention, Formalized: Query, Key & Value
- 5.Multi-Head Attention & the Full Transformer Block
- 6.Decoder-Only vs. Encoder-Decoder: GPT vs. BERT vs. T5
- 7.Pretraining at Scale: Self-Supervised Learning
- 8.Scaling Laws
- 9.Fine-Tuning & RLHF: From GPT-3 to ChatGPT
- 10.Context Windows, Quadratic Attention & Honest Limitations
- 11.Capstone: Tracing a Prompt Through a Real LLM, End to End
Read as One Page
PDF Downloads
Exercise Solutions
llm1-1_exercise1.txtllm1-1_exercise2.txtllm1-1_exercise3.txtllm1-2_exercise1.txtllm1-2_exercise2.txtllm1-2_exercise3.txtllm1-3_exercise1.txtllm1-3_exercise2.txtllm1-3_exercise3.txtllm1-4_exercise1.txtllm1-4_exercise2.txtllm1-4_exercise3.txtllm1-5_exercise1.txtllm1-5_exercise2.txtllm1-5_exercise3.txtllm1-6_exercise1.txtllm1-6_exercise2.txtllm1-6_exercise3.txtllm1-7_exercise1.txtllm1-7_exercise2.txtllm1-7_exercise3.txtllm1-8_exercise1.txtllm1-8_exercise2.txtllm1-8_exercise3.txtllm1-9_exercise1.txtllm1-9_exercise2.txtllm1-9_exercise3.txtllm1-10_exercise1.txtllm1-10_exercise2.txtllm1-10_exercise3.txtllm1-11_exercise1.txtllm1-11_exercise2.txtllm1-11_exercise3.txt