Video links

The lecture recordings of the courses in the course links, arranged by chapter. The reading plan gives the matching slides, notes, and readings. CMU 11-711 publishes slides and code but no public recordings, so it appears only in the reading plan.

Video or playlist	Link	Notes
Stanford CS224N, Spring 2024	Playlist	The numbers below are lecture numbers; for lecture 9 the table links the 2023 recording.
Stanford CS336, Spring 2025	Playlist	The numbers below are lecture numbers. The Spring 2026 recordings update the same sequence.
Andrej Karpathy, Neural Networks: Zero to Hero	Playlist	Code-along lectures that build a language model from counting bigrams to a GPT, and a tokenizer.

By chapter

Chapter	CS224N Spring 2024 lectures	CS336 Spring 2025 lectures	Other recordings
1. Text, Tokens, and Tokenization	—	1: Overview, tokenization	Karpathy: Let’s build the GPT tokenizer
2. N-gram Language Models and Perplexity	5: Recurrent neural networks, first half	—	Karpathy: The spelled-out intro to language modeling
3. Word Embeddings	1: Intro and word vectors · 2: Word vectors and language models	—	—
4. Transformer Language Models	—	2: PyTorch, resource accounting · 3: Architectures, hyperparameters · 4: Mixture of experts	Karpathy: Let’s build GPT, from scratch, in code, spelled out
5. Pretraining and Transfer	9: Pretraining (2023 recording)	—	—
6. Pretraining Data	—	13: Data · 14: Data, filtering and deduplication	—
7. Scaling Laws	—	9: Scaling laws · 11: Scaling laws, details	—
8. Decoding and Text Generation	—	10: Inference	—
9. In-Context Learning and Prompting	10: Post-training, first part	—	—
10. Fine-Tuning and Parameter-Efficient Adaptation	12: Efficient training, section on parameter-efficient fine-tuning	—	—
11. Learning from Human Preferences	10: Post-training, second part · 15: Life after DPO (Nathan Lambert)	15: Alignment, SFT and RLHF	—
12. Reasoning and Test-Time Compute	14: Reasoning and agents, first half	16: Alignment, reinforcement learning from verifiable rewards · 17: Alignment, policy gradients for language models	—
13. Retrieval, Tools, and Agents	14: Reasoning and agents, second half	—	—
14. Evaluating Language Models	11: Benchmarking and evaluation (Yann Dubois)	12: Evaluation	—
15. Machine Translation and Multilingual Models (optional)	6: Sequence-to-sequence models and machine translation	—	—
16. Syntactic Parsing (optional)	4: Dependency parsing · 16: ConvNets, tree recursive networks, and constituency parsing	—	—

The CS224N lectures on neural networks, backpropagation, recurrent networks, attention, and transformers (lectures 3, the second half of 5, and 8) and the CS336 lectures on GPUs, kernels, and parallelism (5–8) review material covered in Deep Learning and are not listed above.