Evaluating understanding tasks - ChatGPT Tutorial

Contents

• Introduction Introduction

  • Quick guide to large language models: Introduction

    2m 41s
• 1. Overview of Large Language Models (LLMs) 1. Overview of Large Language Models (LLMs)

  • (Locked)
    Module 1: Introduction to large language models

    2m 42s
  • (Locked)
    Topics

    48s
  • What are language models?

    22m 51s
  • (Locked)
    Popular modern LLMs

    20m 45s
  • (Locked)
    Applications of LLMs

    3m 36s
• 2. Semantic Search with LLMs 2. Semantic Search with LLMs

  • (Locked)
    Topics

    37s
  • (Locked)
    Introduction to semantic search

    15m 49s
  • (Locked)
    Building a semantic search system

    25m 49s
  • (Locked)
    Optimizing semantic search with cross-encoders and fine-tuning

    28m 58s
• 3. First Steps with Prompt Engineering 3. First Steps with Prompt Engineering

  • (Locked)
    Topics

    42s
  • (Locked)
    Introduction to prompt engineering

    27m 36s
  • Working with prompts across models

    4m 43s
  • (Locked)
    Building a retrieval-augmented generation bot with ChatGPT and GPT-4o

    32m 43s
• 4. Retrieval-Augmented Generation and AI Agents 4. Retrieval-Augmented Generation and AI Agents

  • (Locked)
    Topics

    38s
  • (Locked)
    Introduction to retrieval-augmented generation (RAG)

    14m 1s
  • (Locked)
    Building a RAG bot

    18m 45s
  • (Locked)
    Using open-source models with RAG

    17m 42s
  • (Locked)
    Expanding into AI agents

    19m 24s
• 5. Optimizing LLMs with Fine-Tuning 5. Optimizing LLMs with Fine-Tuning

  • (Locked)
    Module 2: Getting the most out of LLMs introduction

    2m 30s
  • (Locked)
    Topics

    43s
  • (Locked)
    Transfer learning: A primer

    6m 6s
  • (Locked)
    The OpenAI fine-tuning API

    5m 47s
  • (Locked)
    Case Study: Predicting with Amazon Reviews: Part 1

    11m 4s
  • (Locked)
    Case Study: Predicting with Amazon Reviews: Part 2

    37m 38s
• 6. Advanced Prompt Engineering 6. Advanced Prompt Engineering

  • (Locked)
    Topics

    27s
  • (Locked)
    Input/output validation

    9m 28s
  • (Locked)
    Batch prompting and prompt chaining

    7m 57s
  • (Locked)
    Chain-of-thought prompting

    14m 9s
  • (Locked)
    Preventing prompt injection attacks

    5m 9s
  • (Locked)
    Assessing an LLM's encoded knowledge level

    5m 3s
• 7. Customizing Embeddings and Model Architectures 7. Customizing Embeddings and Model Architectures

  • (Locked)
    Topics

    40s
  • (Locked)
    Case study: Building an anime recommendation system

    9m 20s
  • (Locked)
    Using OpenAI’s embedding models

    22m 4s
  • (Locked)
    Fine-tuning an embedding model to capture user behavior

    28m 59s
• 8. AI Alignment: First Principles 8. AI Alignment: First Principles

  • (Locked)
    Topics

    46s
  • (Locked)
    Introduction to AI alignment

    16m 25s
  • (Locked)
    Evaluating alignment plus ethics

    22m 20s
• 9. Moving Beyond Foundation Models 9. Moving Beyond Foundation Models

  • (Locked)
    Module 3: Advanced LLM usage introduction

    3m 22s
  • (Locked)
    Topics

    45s
  • (Locked)
    The vision transformer

    2m 33s
  • (Locked)
    Using cross attention to mix data modalities

    3m 16s
  • (Locked)
    Case study: Visual QA—Setting up a model

    20m 41s
  • (Locked)
    Case study: Visual QA—Setting up parameters and data

    18m 56s
  • (Locked)
    Introduction to reinforcement learning from feedback

    12m 46s
  • (Locked)
    Aligning FLAN-T5 with reinforcement learning from feedback

    21m 37s
• 10. Advanced Open-Source LLM Fine-Tuning 10. Advanced Open-Source LLM Fine-Tuning

  • (Locked)
    Topics

    46s
  • (Locked)
    BERT for multilabel classification: Part 1

    12m 36s
  • (Locked)
    BERT for multilabel classification: Part 2

    26m 59s
  • (Locked)
    Writing LaTeX with GPT-2

    21m 31s
  • (Locked)
    Case study: Sinan’s attempt at wise yet engaging responses—SAWYER

    23m 54s
  • (Locked)
    Instruction alignment of LLMs: Supervised fine-tuning

    28m 23s
  • (Locked)
    Instruction alignment of LLMs: Reward modeling

    21m 49s
  • (Locked)
    Instruction alignment of LLMs: RLHF

    26m 54s
  • (Locked)
    Instruction alignment of LLMs: Using an instruction-aligned LLM

    19m 10s
• 11. Moving LLMs into Production 11. Moving LLMs into Production

  • (Locked)
    Topics

    34s
  • (Locked)
    Cost projecting and deploying LLMs to production

    10m 12s
  • (Locked)
    Knowledge distillation

    24m 4s
• 12. LLM Evaluations 12. LLM Evaluations

  • (Locked)
    Topics

    36s
  • (Locked)
    Evaluating generative tasks: Part 1

    14m 37s
  • (Locked)
    Evaluating generative tasks: Part 2

    18m 45s
  • (Locked)
    Evaluating understanding tasks

    24m 11s
  • (Locked)
    Probing LLMs for world model

    11m 51s
• Conclusion Conclusion

  • (Locked)
    Quick quide to large language models: Summary

    3m 19s