MICRO-COURSE

LLM Internals

LLM Internals: 13 short lessons, one for each concept. Tokens, tokenization, embeddings, positional representations, self-attention, encoder and decoder architectures, context windows, pretraining objectives, next-token prediction, mixture-of-experts models, decoding, sampling, and inference.

One focused topic Learn at your pace

Open topic guide Request a session

Learn this micro-course here. Explore the full course whenever you choose.

RECORDED WALKTHROUGHLLM InternalsNo recording published yet

01 / WATCH

Recorded video

The recording for this topic is not available yet. You can use the written guide below or request a recorded walkthrough.

Request this recording

02 / UNDERSTAND

Start with the topic guide.

Tokens, tokenization, embeddings, positional representations, self-attention, encoder and decoder architectures, context windows, pretraining objectives, next-token prediction, mixture-of-experts models, decoding, sampling, and inference.

This is a learning outline. A complete worked tutorial has not been published for this topic yet.

Concept scope

03 / TRY IT

Prepare a focused example

Choose one question about LLM Internals that you want to answer.
Write down a small example, its inputs or starting point, and the result you expect.
Explain your reasoning and note what you would like reviewed in a live or 1:1 session.

Keep your work

A short example, reasoning and an open question

Check your understanding

Explain the concept in your own words and identify where it fits in the subject.
Show one small example, the expected result and the reasoning behind it.
Record one limitation or edge case and a question to discuss with a mentor.

04 / LEARN WITH HELP

Need a walkthrough?
Stay with this topic.

Request a live session or focused one-to-one guidance on LLM Internals. Bring your questions, a practice attempt or a specific blocker.

Your request stays specific to this micro-course.
Share your preferred schedule in the form.
Availability, timing and fees are confirmed before booking.

Prefer a quick chat?Ask on WhatsApp

CONTINUE WITH ONE CONCEPT

Explore 13 focused topics.

Each topic opens its own micro-course, with its learning guide, recording status and session requests.

13 results1–12

Lesson

Tokens

A short lesson on tokens from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Tokenization

A short lesson on tokenization from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Embeddings

A short lesson on embeddings from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Positional representations

A short lesson on positional representations from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Self-attention

A short lesson on self-attention from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Encoder and decoder architectures

A short lesson on encoder and decoder architectures from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Context windows

A short lesson on context windows from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Pretraining objectives

A short lesson on pretraining objectives from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Next-token prediction

A short lesson on next-token prediction from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson

Lesson

Mixture-of-experts models

A short lesson on mixture-of-experts models from the LLM Internals micro course: what it is, where it shows up in real systems, and a practical example you can apply.

Open lesson