Repolode
Submit
afshinea

afshinea/stanford-cme-295-transformers-large-language-models

A VIP cheatsheet and study guide condensing Stanford's CME 295 course on Transformers and large language models into one place.

View on GitHub
Stars
4,967
Forks
722
Open issues
8
Last commit
last month

What is it?

This repository collects the core concepts from Stanford's CME 295 course on Transformers and Large Language Models into a single cheatsheet. It covers Transformer fundamentals such as self-attention and architecture variants, then moves to LLM topics including prompting, finetuning methods like SFT and LoRA, and preference tuning with RLHF and DPO.

The material also spans reasoning-oriented training (RLVR, OPD), optimizations like distributed training, KV caching and speculative decoding, and applications such as AI agents, evaluation and diffusion LLMs. The content is offered as a PDF cheatsheet, available in many languages, and summarizes a larger companion book with several hundred illustrations.

Key features

  • 01Covers Transformer fundamentals including self-attention and architecture variants
  • 02Explains LLM finetuning with SFT and LoRA and preference tuning with RLHF and DPO
  • 03Summarizes reasoning training methods such as RLVR and OPD
  • 04Lists inference optimizations including KV caching and speculative decoding
  • 05Available as a downloadable PDF cheatsheet in many languages

Use cases

  • Reviewing core LLM and Transformer concepts before interviews or exams
  • Getting a fast overview of modern LLM training pipelines
  • Studying alongside or as a companion to the CME 295 course

Topics

#transformers#llm#deep-learning#fine-tuning#study-guide#cheatsheet#attention-mechanism

FAQ

What is stanford-cme-295-transformers-large-language-models?
A VIP cheatsheet and study guide condensing Stanford's CME 295 course on Transformers and large language models into one place. This repository collects the core concepts from Stanford's CME 295 course on Transformers and Large Language Models into a single cheatsheet. It covers Transformer fundamentals such as self-attention and architecture variants, then moves to LLM topics including prompting, finetuning methods like SFT and LoRA, and preference tuning with RLHF and DPO.
Is stanford-cme-295-transformers-large-language-models free to use? What license does it have?
stanford-cme-295-transformers-large-language-models is open source under the MIT license. Check the license terms before using it in commercial products.
Is stanford-cme-295-transformers-large-language-models still maintained?
Updated within 6 months. The last commit to stanford-cme-295-transformers-large-language-models was on Sep 13, 2026.
What can I use stanford-cme-295-transformers-large-language-models for?
Reviewing core LLM and Transformer concepts before interviews or exams Getting a fast overview of modern LLM training pipelines Studying alongside or as a companion to the CME 295 course
Where can I get stanford-cme-295-transformers-large-language-models?
The source code and installation instructions are on GitHub: https://github.com/afshinea/stanford-cme-295-transformers-large-language-models

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects