К содержимому
learnspaceYOUR NEXT CHAPTER
ПРОСТРАНСТВО ОБУЧЕНИЯ
ГлавнаяКаталог курсовМоё обучениеCoursera

Знания без границ

Учитесь у лучших университетов и компаний мира.

Открыть Coursera
Интеграция
Пространство университета
Моё пространствоСтраница курса
↵
ЯЛичный кабинетСтудент
© 2026 LearnSpaceКаждый день — возможность узнать больше.Помощь
Generative Pre-trained Transformers (GPT) · LearnSpace
Назад в каталог
courseraАнализ данных

Generative Pre-trained Transformers (GPT)

Курс от University of Glasgow
Средний≈ 13.3 чАнглийский
О курсеНавыкиПрограммаПреподаватели

О курсе

Large Language Models (LLMs), including GPT models that power applications such as ChatGPT, are transforming how people interact with technology and how computers understand and generate language. In this course, you'll explore the core concepts of natural language processing (NLP) and language modelling that underpin today's generative AI systems. You'll learn how language models are trained, how Transformer architectures revolutionised modern AI, and why they have become the foundation for a wide range of applications, from conversational assistants and content generation to summarisation, translation, and question answering. Along the way, you'll examine the strengths and limitations of LLMs, including topics such as hallucinations, evaluation, responsible AI, and the ethical considerations involved in developing and deploying these technologies. Through hands-on Python labs, you'll explore the building blocks of Transformer-based language models, experiment with text generation, and gain practical experience applying smaller language models to real-world tasks. Regular practice quizzes and interactive learning activities will reinforce key concepts and help prepare you for the graded assessments. Whether you're looking to understand how modern LLMs work or build a foundation for working with generative AI, this course provides the knowledge and practical experience to get started.

Навыки, которые вы освоите

Large Language ModelingRisk Management FrameworkGenerative Model ArchitecturesResponsible AIEmbeddingsData EthicsModel TrainingLLM ApplicationChatGPTNatural Language ProcessingModel EvaluationGenerative AI

Программа курса

3 модулей · 66 учебных материалов

01Language Modeling15 материалов

N-Gram Language Models

Course and Instructor IntroductionsВидеоStay connected with UofG OnlineЧтениеWhat is a Language Model?ВидеоN-gram Language ModelsВидео

Учитесь у экспертов

Mary Ellen Foster

Dr

Sean MacAvaney

Преподаватель курса

Jake Lever

Dr

Generative Pre-trained Transformers (GPT)
В каталоге вашей программы

Инвестируйте в себя

Новые знания — в удобное для вас время.

Начать на Coursera

Обучение откроется на Coursera
в новой вкладке

Обучение на Coursera

≈ 13.3 ч

3 модулей

Язык: Английский

Субтитры: Арабский, Французский, Узбекский, Украинский, Китайский (Китай), Греческий, Итальянский, Бразильский португальский, Нидерландский, Корейский, Немецкий, Русский, Тайский, Индонезийский, Шведский, Турецкий, Испанский, Хинди, Японский, Венгерский, Польский

Часть программы вашего университета
Building Your Own N-Gram Language ModelЛабораторная
Module 1 Lesson 1 Practice QuizЗадание

Evaluating Language Models

Evaluating Language ModelsВидеоCalculating Perplexity from an N-Gram Language ModelЛабораторнаяModule 1 Lesson 2 Practice QuizЗадание

Text Generation with Language Models

Generating TextВидеоBuilding a Simple Text GeneratorЛабораторнаяModule 1 Lesson 3 Practice QuizЗаданиеSummary of ModuleВидеоOptional ReadingЧтениеModule 1 QuizЗадание
02Transformers and GPT19 материалов

Tokenization

Introduction to the ModuleВидеоRepresenting Text with Numerical Vectors​ВидеоWords, Tokens or Sub-tokens​ВидеоTokenizationЛабораторнаяModule 2 Lesson 1 Practice QuizЗадание

Neural language models & Transformers

Neural Language ModelsВидеоThe Transformer ArchitectureВидеоHow are Transformers Trained?ВидеоLanguage Modelling with TransformersЛабораторнаяCalculating PerplexityЛабораторнаяModule 2 Lesson 2 Practice QuizЗадание

GPT

GPT and BERTВидеоUsing GPTВидеоThe Ethical Challenges of Large Language ModelsВидеоLanguage Generation with TransformersЛабораторнаяOn the Dangers of Stochastic ParrotsЧтениеModule 2 Lesson 3 Practice QuizЗаданиеWrap-up of the Module
03Applications and Implications32 материалов

Hallucinations, Reliability, and Risk

Introduction to the ModuleВидеоWhat are LLM Hallucinations and Why Do They Occur?ВидеоWhy LLM Hallucinations Matter: Risk and ConsequencesВидеоOptional: ChatGPT Lawyer detailsЧтениеModule 3 Lesson 1 Practice QuizЗаданиеModule 3 Lesson 1 Dialogue: Investigating GPT Hallucinations in PracticeDIALOGUE

Real-World Use Cases: Chatbots and Academic Uses

Appropriate Academic Use of LLMsВидеоAcademic Risks: Hallucinations, Detection, and IntegrityВидеоChatbots and the Illusion of UnderstandingВидеоWhy Reinforcement Learning with Human Feedback Improves ChatbotsВидеоRequired: OpenAI Blog Post about ChatGPT and RLHFЧтениеRequired: Research about AI Detectors and Non-native SpeakersЧтение

Documented and Speculative Risks

Creativity and Copyright (1)ВидеоCreativity and Copyright (2)ВидеоRisks and RegulationsВидеоRequired: Scientific American Article on AI RisksЧтениеRequired: UNESCO Report on AI and CultureЧтениеOptional: AI-generated E-books and JournalismЧтение

Pulling it all together

Module 3 Final QuizЗаданиеCourse RecapВидеоKeeping Learning with UofG OnlineЧтение
Видео
Module 2 QuizЗадание
Optional: Ouyang et al. (2022) Paper about InstructGPTЧтение
Optional: News Stories about the Humans in RLHFЧтение
Optional: Russell Group and University of Glasgow Principles on AI in EducationЧтение
Module 3 Lesson 2 Practice QuizЗадание
Module 3 Lesson 2 Dialogue: How Chatbots Shape Trust and Decision‑MakingDIALOGUE
Optional: Generative AI, Copyright, and Creative IndustriesЧтение
Optional: Open LettersЧтение
Optional: Approaches to AI RegulationЧтение
Optional: One Hundred Year Study on Artificial IntelligenceЧтение
Module 3 Lesson 3 Practice QuizЗадание
Module 3 Lesson 3 Dialogue: How Organisations Respond to Generative AIDIALOGUE