Fine-Tune Llama 3.2 Model on Custom Dataset - Easy Step-by-Step Tutorial

This video shows how to easily fine-tune Llama 3.2 1B model for your phone or laptop or any other custom dataset on free google colab using Unsloth.
🔥 Buy Me a Coffee to support the channel: ko-fi.com/fahd...
🔥 Get 50% Discount on any A6000 or A5000 GPU rental, use following link and coupon:
bit.ly/fahd-mirza
Coupon code: FahdMirza
▶ Become a Patron 🔥 - / fahdmirza
#llama32 #llama1b #llama3b #llama11b #llama90b #unsloth
PLEASE FOLLOW ME:
▶ LinkedIn: / fahdmirza
▶ RUclips: / @fahdmirza
▶ Blog: www.fahdmirza.com
RELATED VIDEOS:
▶ Resource colab.research...
All rights reserved © Fahd Mirza

[@fahdmirza]

🔥Agentic RAG with Llama 3.2 on Custom Dataset - ruclips.net/video/BsfrI0lMLjY/видео.htmlsi=sMdFErOxUKtc6Qn4
🔥Enters Llama 3.2 with Text and Vision in 1B, 3B, 11B, and 90B, ruclips.net/video/SfjQCHsZ6Ec/видео.html
🔥Install Llama 3.2 1B Instruct Locally - Multilingual On-Device AI Model, ruclips.net/video/aKEUjAjJY7Q/видео.html
🔥Llama 3.2 3B Instruct - Small Yet Powerful Meta Model - Install Locally, ruclips.net/video/xTgyrC-HZ7o/видео.html
🔥Llama-3.2 11B Vision Instruct - Install Locally - ruclips.net/video/BlLtQfx6I_o/видео.htmlsi=jo3MaqZCVD1rW3Cv

[@Shivam-bi5uo]

next video please cover on how we could fine tune the 3.2 vision models, fahd.

[@ritpop]

Do you think is worrhy trying to train a rp model with the 1 or 3b mosels or wait a little more to the community do it?

[@SaiyD]

Thanks Fahd Mirza, you re awesome!

[@fahdmirza]

Thank you too

[@JeganJegatheesh]

Thanks Fahd Mirza. Can I use this saved model in ollama?

[@tayyabshahzad10]

Great work. Today I was talking with my colleague about how we can finetune (particularly the vision models also make one video about them as well) these models for specific tasks and here we go. 😍😍❤️❤️

[@fahdmirza]

Great suggestion!

[@atultiwari88]

thank you for sharing this. But how can I convert the csv data having just questions & answers into sharegpt format. And more importantly if the dataset is just questions and answers, then why should we convert it to sharegpt format, why cant we just use alpaca format? thank you

[@fahdmirza]

I have already done few videos on it, please search the channel.

[@Piotr_Sikora]

Nice job, but better will be train guard model for different languages.

[@fahdmirza]

Good point!

[@amemon827]

kindly let me know the transformer version dependency

[@fahdmirza]

go with the latest one.

[@razdingz]

bro nice simplification , this is good yaya

[@fahdmirza]

Thank you so much 🙂