# lordog/dive-into-llms architecture

> Architecture diagram of the GitHub repository lordog/dive-into-llms, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-28). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/lordog/dive-into-llms
- Repository: https://github.com/lordog/dive-into-llms

## Overview

This repository is a free, hands-on educational tutorial series, not an application with a shared runtime. Learners use each chapter’s guide and notebook to explore a topic and carry out its practical exercises. The chapters cover model adaptation and prompting, editing and reasoning, multimodal and agent capabilities, and safety and alignment. The README also documents a separate Ascend-focused course hosted by the Huawei community. Chapter relationships below reflect the documented curriculum, not inferred calls between notebooks; notebook internals were not sampled.

Main entry points: [`documents/chapter1/README.md`](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter1/README.md) (Fine-tuning and deployment), [`documents/chapter2/README.md`](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter2/README.md) (Prompting and reasoning), [`documents/chapter3/README.md`](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter3/README.md) (Knowledge editing), [`documents/chapter4/README.md`](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter4/README.md) (Math reasoning)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_foundations["Model practice"]
  node_tuning["Fine-tuning and deployment<br/>[README.md]"]
  node_prompting["Prompting and reasoning<br/>[README.md]"]
  node_editing["Knowledge editing<br/>[README.md]"]
  node_math["Math reasoning<br/>[README.md]"]
end

subgraph group_capabilities["Model capabilities"]
  node_watermark["Model watermarking<br/>[README.md]"]
  node_steganography["LLM steganography<br/>[README.md]"]
  node_multimodal["Multimodal models<br/>[README.md]"]
  node_guiagent["GUI agents<br/>[README.md]"]
end

subgraph group_safety["Safety and alignment"]
  node_jailbreak["Jailbreak attacks<br/>[README.md]"]
  node_agentsafety["Agent safety<br/>[README.md]"]
  node_rlhf["RLHF safety alignment<br/>[README.md]"]
end

node_learner(("Learner"))
node_ascend["Ascend development course"]

node_learner -->|"studies and practices"| node_tuning
node_learner -->|"studies and practices"| node_prompting
node_learner -->|"studies and practices"| node_editing
node_learner -->|"studies and practices"| node_math
node_learner -->|"studies and practices"| node_watermark
node_learner -->|"studies and practices"| node_jailbreak
node_learner -->|"studies and practices"| node_steganography
node_learner -->|"studies and practices"| node_multimodal
node_learner -->|"studies and practices"| node_guiagent
node_learner -->|"studies and practices"| node_agentsafety
node_learner -->|"studies and practices"| node_rlhf
node_learner -.->|"accesses course"| node_ascend

click node_tuning "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter1/README.md"
click node_prompting "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter2/README.md"
click node_editing "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter3/README.md"
click node_math "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter4/README.md"
click node_watermark "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter5/README.md"
click node_jailbreak "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter6/README.md"
click node_steganography "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter7/README.md"
click node_multimodal "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter8/README.md"
click node_guiagent "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter9/README.md"
click node_agentsafety "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter10/README.md"
click node_rlhf "https://github.com/lordog/dive-into-llms/blob/main/documents/chapter11/README.md"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_tuning,node_prompting,node_editing,node_math toneBlue
class node_watermark,node_steganography,node_multimodal,node_guiagent toneAmber
class node_jailbreak,node_agentsafety,node_rlhf toneMint
class node_learner,node_ascend toneIndigo
```

## Components

### Model practice

- [Fine-tuning and deployment](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter1/README.md): `documents/chapter1/README.md`
- [Prompting and reasoning](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter2/README.md): `documents/chapter2/README.md`
- [Knowledge editing](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter3/README.md): `documents/chapter3/README.md`
- [Math reasoning](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter4/README.md): `documents/chapter4/README.md`

### Model capabilities

- [Model watermarking](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter5/README.md): `documents/chapter5/README.md`
- [LLM steganography](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter7/README.md): `documents/chapter7/README.md`
- [Multimodal models](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter8/README.md): `documents/chapter8/README.md`
- [GUI agents](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter9/README.md): `documents/chapter9/README.md`

### Safety and alignment

- [Jailbreak attacks](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter6/README.md): `documents/chapter6/README.md`
- [Agent safety](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter10/README.md): `documents/chapter10/README.md`
- [RLHF safety alignment](https://github.com/lordog/dive-into-llms/blob/main/documents/chapter11/README.md): `documents/chapter11/README.md`

### Other components

- **Learner**
- **Ascend development course**

## Connections

- Learner → Fine-tuning and deployment: studies and practices
- Learner → Prompting and reasoning: studies and practices
- Learner → Knowledge editing: studies and practices
- Learner → Math reasoning: studies and practices
- Learner → Model watermarking: studies and practices
- Learner → Jailbreak attacks: studies and practices
- Learner → LLM steganography: studies and practices
- Learner → Multimodal models: studies and practices
- Learner → GUI agents: studies and practices
- Learner → Agent safety: studies and practices
- Learner → RLHF safety alignment: studies and practices
- Learner → Ascend development course: accesses course