What you'll learn:
- Explore the fundamentals of LLM and LLM Fine tuning including - Why Fine tuning is needed, How it works, Applications, etc.
- Learn the complete workflow, architecture, essential steps involved for Fine tuning Large language models (LLMs).
- Different types of LLM Fine tuning techniques including RLHF, PEFT, LoRA, QLoRA, Standard, Sequential and Instructions based LLM Fine tuning.
- Get a comprehensive walkthrough of OpenAI Dashboard and Playground to get a holistic understanding of wide range of tools that OpenAI offers for Generative AI.
- Prepare the dataset as per OpenAI accepted JSONL format for Fine-tuning GPT models, while also calculating the fine-tuning cost in advance.
- Learn step-by-step through Hands-on sessions on how to fine tune OpenAI's GPT model on custom dataset using Python.
- Evaluate and compare the fine tune model with the base pre-trained GPT model to to assess improvements in accuracy and performance.
- Discover Best practices for LLM fine-tuning.
"Large Language Models (LLMs) have revolutionized the AI industry, providing unprecedented precision, expanding the possibilities of artificial intelligence. However the pre-trained LLMs may not always meet the specific requirements of an organization, hence there is always a need to Fine tune the LLMs to tailor these models to your unique tasks and requirements."
This comprehensive course is designed to equip you with the skills of LLM Fine tuning technique. It starts with a thorough introduction to the fundamentals of fine tuning while highlighting its critical role to make the LLM models adapt to your specific data. Then we will dive into Hands-on sessions covering the entire Fine-tuning workflow to Fine tune OpenAIGPTmodel. Through practical sessions, you'll step-by-step learn to prepare & format datasets, execute OpenAI fine tuning processes, and evaluate the model outcomes.
By the end of this course, you will be proficient in Fine tuning the OpenAI's GPT model to meet specific organizational needs and start your career journey as LLM Fine tuning engineer.
____________________________________________________________________________________________
What in nutshell is included in the course ?
[Theory]
We'll start with LLM and LLMFine tuning's core basics and fundamentals.
Discuss Why is Fine tuning needed, How it works, Workflow of Fine tuning and the Steps involved in it.
Different types of LLM Fine tuning techniques including RLHF, PEFT, LoRA, QLoRA, Standard, Sequential and Instructions based LLM Fine tuning.
Best practices for LLMFine tuning.
[Practicals]
Get a detailed walkthrough of OpenAI Dashboard and Playground to get a holistic understanding of wide range of tools that OpenAI offers for Generative AI.
Follow the OpenAIFine tuning workflow in practical sessions covering Exploratory Data Analysis (EDA), Data preprocessing, Data formatting, Creating fine tuning job, Evaluation.
Understand the OpenAI specialized JSONL format that it accepts for training & test data, and learn about 3 important roles - System, User, Assistant.
Calculate the Token count and Fine tuning cost in advance using Tiktoken library.
Gain Hands-on experience in fine tuning OpenAI's GPT model on a custom dataset using Python through step-by-step practical sessions.
Assess the accuracy and performance of the fine-tuned model compared to the base pre-trained model to evaluate the impact of fine-tuning.
