Scribd
Upload
22 views

nAI Program - LLMs Curriculum

The course covers topics related to large language models including getting started with LLMs, building applications using prompt engineering, building production ready question answering systems, finetuning LLMs, and training LLMs from scratch. The instructors are experts from Analytics Vidhya, HuggingFace, and LlamaIndex.

Uploaded by

VijayKhare
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
22 views

nAI Program - LLMs Curriculum

The course covers topics related to large language models including getting started with LLMs, building applications using prompt engineering, building production ready question answering systems, finetuning LLMs, and training LLMs from scratch. The instructors are experts from Analytics Vidhya, HuggingFace, and LlamaIndex.

Uploaded by

VijayKhare
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

GenAI Program: LLMs Courses

1. Getting Starting with LLMs


2. Building LLM applications using Prompt Engineering
3. Building Production Ready RAG systems using LlamaIndex
4. Finetuning LLMs
5. Training LLMs from Scratch

Instructors:

Kunal Jain, Founder and CEO of Analytics Vidhya


Sourab, ML Engineer at HuggingFace
Ravi Theja, Developer Advocate at LlamaIndex

Getting started with LLMs

Module No. Module Name Lesson Name


1 Evolution of NLP Evolution of NLP: Pre Deep Learning
Evolution of NLP: Deep Learning
Tranformers
2 Introduction to LLMs What is a language model?
What is a large language model?
Understanding Foundational Model
Different types of LLMs: Continuing the text
Different types of LLMs: Instruction Tuned
Encoder LLMs, Decoder LLMs and Encoder
Decoder LLMs
3 The Current State of the Art in LLMs The Current State of the Art in LLMs
What Next?

Building Production Ready RAG systems using LlamaIndex

Module
No. Module Name Item/Lesson Name
1 Introduction to RAG systems Why RAG systems?
What is a RAG system?
A brief overview of steps involved in a RAG Framework
Quiz
2 Introduction to LlamaIndex What is a LlamaIndex?
Why LlamaIndex?
Quiz
3 Components of LlamaIndex Overview of Different Components of LlamaIndex
Data Sources
Indexing
Embeddings
VectorDBs
Quiz
LLMs
Retrieval
Decoupling chunks used for retrieval vs. chunks used
for synthesis
Adding Metadata
Quiz
Code:Build your first QA system on
4 private data
5 Evaluation of RAG systems Necessity of Evaluation
Retrieval Evaluation
Response Evaluation
Quiz
6 Code: RAG + Evaluation
Advanced approaches for powerful
7 RAG system Router Query Engine, SubQuestionQuery Engine
Structured Retrieval for Larger Document
Sets:Metadata filters + Auto Retrieval
Document Hierarchies + Recursive Retrieval
Finetuning Embeddings
Finetuning with Retrieval Augmentation (RADIT)
8 Agents
Building a powerful RAG system
9 using LlamaIndex

Finetuning LLMs

Module Module Name Item/Lesson Name


No.
1 Introduction to Finetuning LLMs What is Finetuning LLMs?
Why Finetuning LLMs?
Research Papers and blogs for additional
details
Quiz
2 Understanding different finetuning techniques Feature Extraction
Retraining all parameters
The problem with traditional methods
Quiz
Setting up the runpod instance
Code Part1: Finetuning NLP usecase using
feature extraction and full finetuning
Code Part2: Finetuning NLP usecase using
feature extraction and full finetuning
Research Papers and blogs for additional
details
Introduction to Parameter Efficient Finetuning What is Parameter Efficient Finetuning
3 Techniques Techniques?
How PEFT addresses the issues with Full
Fine-tuning
Different types of PEFT methods
Quiz
4 Prompt Tuning Techniques Prompt Tuning
Prefix Tuning
Code:Finetuning NLP usecase using Soft
Prompting Techniques
Quiz
5 Adaptor Modules (IA)3
LoRA
Quantization
QLoRA
AdaLoRA
Scaling and combining multiple loras
Code:Finetuning NLP usecase using PEFT
LoRA
Code: Finetuning instruction following LLM
using PEFT

Training LLMs from Scratch

Module No. Module Name Item/Lesson Name


1 Training LLMs from Scratch Scaling laws
Parallel and Distributed
computing strategies
Challenges involved in
training LLMs
Best practices to train
LLMs
How do you train LLMs
from scratch?
Project: Continuing text
Project: Instruct
optimized + RLHF
2 Evaluating LLMs

Building LLM App using Prompt Engineering

Module
No. Module Name Lesson Name
Overview of different methods to build LLM
1 How to build LLM Applications? applications
2 Introduction to Prompt Engineering Why do we need Prompt Engineering?
What is Prompt Engineering?
Prompt Engineering using ChatGPT API and
3 Open Source LLM Introduction to ChatGPT API
Prompt Engineering with Open Source LLM
Understanding Different Prompt Engineering
5 Techniques Zero-shot Prompting
Few-shot Prompting
CoT Prompting
Self Consistency
ToT
Advanced Prompt Engineering Techniques
6 Building LLM App using Prompt Engineering

You might also like