Video course · 8 chapters · 157 min · certificate
Fine Tuning LLM Models – Generative AI Course
Krish Naik's freeCodeCamp course on fine-tuning LLMs: quantisation intuition (symmetric and asymmetric), LoRA and QLoRA maths, fine-tuning Llama 2 with QLoRA, 1-bit LLMs, fine-tuning Google Gemma, no-code LLM pipelines, and fine-tuning with your own data on Gradient.
What you'll learn
- Explain quantisation and why it speeds inference
- Understand LoRA and QLoRA matrix decomposition
- Fine-tune Llama 2 with QLoRA on custom data
- Describe 1-bit (ternary) LLMs
- Fine-tune Google Gemma with LoRA
- Build no-code pipelines and fine-tune on Gradient
Chapters
8 chapters · 157:05-
1:39
01Intro Members
Introduction
Why fine-tune?
-
32:24
02Quantisation Members
Quantisation intuition
Smaller, faster.
-
22:23
03LoRA Members
LoRA and QLoRA intuition
Low-rank updates.
-
24:09
04Llama 2 Members
Fine-tuning with Llama 2
QLoRA in practice.
-
16:58
051-bit Members
1-bit LLMs
Ternary weights.
-
22:12
06Gemma Members
Fine-tuning Google Gemma
LoRA on Gemma.
-
20:48
07No-code Members
No-code LLM pipelines
Build without code.
-
16:32
08Gradient Members
Fine-tuning with custom data on Gradient
Your own data.
Jobs this course suits
Our AI checked this course against 500 jobs; these get the most out of it. Each job links to its learning path.