LLM 101
A clear primer on large language models: their history from n-grams to the Transformer to ChatGPT, how they work (tokens, attention, context windows), the training pipeline (pretraining, SFT, RLHF, alignment), prompting and RAG, the key companies and people, and the numbers and milestones that define the field.
602 cards10 lessonsFree
LLM 101 · FACT
Pre-Transformer NLP: n-grams
Early language models predicted the next word by counting how often word sequences appeared in text (n-grams). They were fast but couldn't capture meaning across long distances — only the last few words mattered.
Ten of its 602 cards, from the same pack file the app reads.
10 lessons.
Each lesson has a study guide to read and objectives you can test yourself on. Its cards and games show up in your feed.
- 01
A Brief History of LLMs
Before the Transformer · The GPT era · ChatGPT and the modern wave
- 02
How LLMs Work
Tokens, embeddings and attention · How text is generated
- 03
The Training Pipeline
Pretraining to RLHF · Alignment, inference and compute
- 04
Prompting, RAG and Using LLMs
Prompting techniques · RAG, tools and agents
- 05
Companies, People and Models
The companies · The godfathers and key researchers · Founders, leaders and the ecosystem
- 06
Key Numbers and Milestones
Model sizes and dates · Scaling, context and misconceptions
- 07
What LLMs Are Good For
Language and writing tasks · Knowledge, search and code · Work and everyday tasks · High-stakes domains and choosing well
- 08
Limits and Responsible Use
Reliability and staleness · Bias, privacy and your data · Security and societal risks · Limits, safeguards and deployment
- 09
Quick Recall
Core terms
- 10
Key Concepts
Key People · Key Things
- Cards
- 602
- Lessons
- 10
- Version
- 1.1.4
- Language
- English
The whole pack is one file. Read the pack file ↗
The app is on its way to the App Store.
FlashFeed is free, built for iPhone, and needs no account. When it's live, the link will be here.