ConceptAI & ML
Token
Either whole words or subword units, pieces of a word, or it could even be punctuation or numbers or so forth.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Lecturers
2Each in their own terms, from the easiest level.
Phillip Isola
A vector of neurons functioning as an encapsulated packet or bundle of information, serving as the fundamental unit operated on by transformers.
Lectures that assume it
16The first 12 are shown.
- Deep Dive into Large Language Models: Pretraining to RLHF
- Transformer Circuits, Induction Heads, In-Context Learning
- Generative Models: Conditional Models
- Reinforcement Learning from Human Feedback: Algorithms & Applications
- Objective-Driven AI: World Models that Learn, Remember, Reason, and Plan
- Mixture of Experts (MoE) paradigm and the Switch Transformer
- Symbols and Grounding in Large Language Models
- Digital Design and Computer Architecture: Problem Solving I (ETH Zürich, Spring 2025, Lecture 26)
- RL with Sequence Models & Language Models, Part 3
- GRIM-Filter: In-Memory Seed Location Filtering in 3D-Stacked DRAM
- Deep Learning in Biology
- The Future of Programming will be Live
Connections
27The lectures that define it or assume it, and the concepts next to it.
Concepts9
- Graph neural networks
- Language model
- Neural networks
- Vector
- Tokenization
- AI models as neural compute units
Show 3 moreShow fewer
Lectures18
- A Hacker's Guide to Language Models
- Architectures: Transformers
- Deep Dive into Large Language Models: Pretraining to RLHF
- Transformer Circuits, Induction Heads, In-Context Learning
- Generative Models: Conditional Models
- Reinforcement Learning from Human Feedback: Algorithms & Applications
Show 12 moreShow fewer
- Objective-Driven AI: World Models that Learn, Remember, Reason, and Plan
- Mixture of Experts (MoE) paradigm and the Switch Transformer
- Symbols and Grounding in Large Language Models
- Digital Design and Computer Architecture: Problem Solving I (ETH Zürich, Spring 2025, Lecture 26)
- RL with Sequence Models & Language Models, Part 3
- GRIM-Filter: In-Memory Seed Location Filtering in 3D-Stacked DRAM
- Deep Learning in Biology
- The Future of Programming will be Live
- Beyond LLMs: Agents, Emergent Abilities, Intermediate-Guided Reasoning, BabyLM
- Transformers that Transform Well Enough to Support Near-Shallow Architectures
- How Always-On Agents Work
- Overview of Transformers