Fine-Tune Llama 3 Model on Custom Dataset - Step-by-step Tutorial

preview_player
Показать описание
This video is an easy tutorial to fine-tune Llama 3 model on colab or locally using your own custom dataset.

#llama3 #lama3 #metallama3 #llama3400B #llama38b #llama400B

PLEASE FOLLOW ME:

RELATED VIDEOS:

All rights reserved © 2021 Fahd Mirza
Рекомендации по теме
Комментарии
Автор

Absolutely wonderful tutorial as always!

danielhanchen
Автор

I want to develop a chatbot for our technical support team with our own dataset.
I am thinking to use langchain and llama 3.1,
Can you please guide me?

prajwalm.s
Автор

hello I upload my model on huggingface can you tell me how to use it?

MahmoudOmar-upbv
Автор

Many of the tutorials focus on already existing data. How can we generate custom data in the same format based on our own knowledge base ?

speedster
Автор

i want to integrate this llama to a vision encoder to make it a multimodal for vqa how can i do that ?? can you help me??

VIVEKKUMAR-dohq
Автор

sir I want to finetune llama3 for language translation so how can I make a data set for it ( it is English to Marathi)
all the steps will be same as you shown in the video??

Rits-lr
Автор

this tutorial is for ai researcher or having advanced level knowledge not for beginner and medium level researcher

shinwarikhan
Автор

What to do if i want to use my own PDF data instead of ALpaca one. Pdf which dont have columns like Alpaca. All text in PDF.

syedmuzammilahmed
Автор

fantastic video! A new fan here.. it will be great if you use local data.. I would like to see how you do that locally if posible ..

fabsync
Автор

Which loss function did you use for training on custom dataset?

kashifraza
Автор

can we do the sequence classification task in similar manner?

rishavranaut
Автор

Great sir thanks to make this video for us..I have custom dataset which is only having numbers question and answer like this.
1, Question, ?"answer".
2, Question, ?"answer".
.
.
.
3k, Qustion, ?"Ans".
how can I train this please give me instruction your big fan..

hassubalti
Автор

How should I format my custom data and load it

onlineinformation
Автор

Friend, please teach how to create the txt file to upload and train, as well as teach how to prepare the file for refinement. Thanks. Pt-Br Translated

alexsanders
Автор

Can you share this colab notebook? Thanks for sharing

AudibleYuva
Автор

Seems we need new benchmarks for Ai as models get new capabilities all the time

PascalMeienberg
Автор

The conversion to gguf is not working, probably because I'm stupid hahahha

lwll
Автор

this is not custom dataset, this is simply running everything on there, it is not showing how to change dataset to use mine....

Khser