Fine Tuning Large Language Models with InstructLab

  Рет қаралды 16,250

IBM Technology

IBM Technology

Күн бұрын

Download the AI model guide to learn more → ibm.biz/Bdaw7r
Learn more about the technology → ibm.biz/Bdaw7i
Want to get more out of your language models? Follow Cedric Clyburn as he shows how to fine-tune large language models using InstructLab, an open-source tool that can help you customize and specialize them for specific tasks. By fine-tuning your language models, you'll be able to tackle complex tasks with ease, automate repetitive work, and unlock new insights that help you solve real-world problems and achieve your goals.
AI news moves fast. Sign up for a monthly newsletter for AI updates from IBM → ibm.biz/Bdaw7j
#finetuning #largelanguagemodels #ai

Пікірлер: 25
@khairullahhabib3982
@khairullahhabib3982 10 күн бұрын
Very good delivery! I can watch this guy explain stuff all day. Keep it up! I can't believe all this knowledge is just free out here
@cloudnativecedric
@cloudnativecedric 17 күн бұрын
Many thanks for the opportunity!
@murunatan6661
@murunatan6661 18 күн бұрын
This is really good. Easy to understand and implement.
@taygundogan
@taygundogan 17 күн бұрын
Yeeeey Cedric, more videos from Cedric please please!!
@stanTrX
@stanTrX 12 күн бұрын
Thanks. Does it work the same with ollama?
@learnbydoingwithsteven
@learnbydoingwithsteven 18 күн бұрын
Very clear. Gonna try it.
@PB-kx4vv
@PB-kx4vv 12 күн бұрын
InstructLab presentations lead me to fantasize about training a model to shorten the learning curve for large open source projects. For example, the code-aster finite element package, with huge amounts of documentation and many documented test cases can many structural and dynamic and even thermal mechanical systems. However, the combinations of features which work compatibly with each other feels to a beginner like a fractal landscape. It is ok to go through an example, but it is easy to loose footing at near adjacencies. It would be nice to talk to a model about strategies to construct a new model, which can reference particular documents and examples, and identify prospective strategies as self conflicting. But when I imagine mapping this problem to instruct lab, I imagine it to be a more daunting task than just working with the program and gaining experience, and reading a lot.
@AmeerHamza-cy6km
@AmeerHamza-cy6km 14 күн бұрын
how can I train it on PHP programming language, and some php projects.
@aganithshanbhag
@aganithshanbhag 13 күн бұрын
question answer set (vast training material on php programming)
@ml00000
@ml00000 18 күн бұрын
Excellent presenter!
@andrewcameron4172
@andrewcameron4172 14 күн бұрын
What version of ilab were you running in this demo?
@cloudnativecedric
@cloudnativecedric 13 күн бұрын
Ah, so this was InstructLab v.17 when we recorded :)
@gokcerbelgusen1062
@gokcerbelgusen1062 18 күн бұрын
I will try this, thank you
@maneeshs3876
@maneeshs3876 16 күн бұрын
Nice video !
@justwanderin847
@justwanderin847 18 күн бұрын
I was just wondering how they really train AI. This helps.
@ZakinAbdul
@ZakinAbdul 15 күн бұрын
That was awesome, and I was wondering, can we fine-tune that model with an RAG chatbot-like, chat with it and feed it new info through our chats?
@LoVe-iu9rd
@LoVe-iu9rd 3 күн бұрын
May I know what is your laptop spec?
@rajavemula3223
@rajavemula3223 7 күн бұрын
Can fine tuning can be done with cpu? I mean without gpu?
@munawwarkhan1926
@munawwarkhan1926 18 күн бұрын
This is a great video and a good intro to an amazing tool. Just one suggestion, it does need some knowledge and background of computer science and data structures. I don't think it is for people with zero knowledge or background as the video suggestsin the beginning. Amazing content IBM, learning a lot here.
@cloudnativecedric
@cloudnativecedric 17 күн бұрын
Thank you very much for the feedback! That is true, there are some basics that are helpful in doing this, as well as terminal usage skills, but what we're working on as well is a user interface for the upstream InstructLab project, so it's essentially a simple form to include Q&A pairs, source documents, and attribution! Then the rest of the process like data generation and training is automated :)
@Pregidth
@Pregidth 12 күн бұрын
@@cloudnativecedric If I understand correctly, by providing exact Q&A pairs during the fine-tuning process, we are effectively guiding the LLM to produce specific, deterministic answers to certain questions. Does this mean we are reducing the inherent randomness in the answers that LLMs typically generate based on their pre-trained weights? If so, wouldn’t this approach limit the model’s flexibility to incorporate its broader pre-trained knowledge into the context of the fine-tuned domain?
@george_davituri
@george_davituri 18 күн бұрын
impressive, need to try cool stuff
@philtoa334
@philtoa334 18 күн бұрын
Nice.
@jacquesgastebois
@jacquesgastebois 17 күн бұрын
I want to do the same with a tiny model please
@vdpoortensamyn
@vdpoortensamyn 17 күн бұрын
Our Granite models are quite tiny. 😊
RAG vs. Fine Tuning
8:57
IBM Technology
Рет қаралды 92 М.
Large Language Models explained briefly
8:48
3Blue1Brown
Рет қаралды 919 М.
99.9% IMPOSSIBLE
00:24
STORROR
Рет қаралды 31 МЛН
Llama: The Open-Source AI Model that's Changing How We Think About AI
8:46
Large Language Models (LLMs) - Everything You NEED To Know
25:20
Matthew Berman
Рет қаралды 136 М.
From Idea to AI: Building Applications with Generative AI
7:13
IBM Technology
Рет қаралды 15 М.
Fine-tuning Large Language Models (LLMs) | w/ Example Code
28:18
Shaw Talebi
Рет қаралды 375 М.
2024's Biggest Breakthroughs in Computer Science
10:47
Quanta Magazine
Рет қаралды 255 М.
What is the Dark Web? A Guide to the Dark Side of the Internet
9:49
IBM Technology
Рет қаралды 1 МЛН
99.9% IMPOSSIBLE
00:24
STORROR
Рет қаралды 31 МЛН