Practical-LLM-Fine-Tuning

This repository contains hands on tutorials on fine tuning LLMs using different fine tuning techniques

Parameter Efficient Fine Tuning (PEFT)

• PEFT for custom summarization
• Fine tuning Phi3/Phi2 on custom dataset
• Fine tuning custom model using gemini 1.5

Instruction Tuning

• Fine tuning Base LLAMA as Instruct Model
• Fine tuning LLAMA3 using transformers

Full Fine Tuning

• Full fine tuning on BART for summarization

Fine tuning vision models

• Vision transformer fine tuning

Fine tuning quantized models

• Fine tuning Falcon 7B quantized model

Fine tuning for specific use cases

• Fine tuning T5 for custom summarization
• Fine tuning roberta base for financial sentiment classification
• Fine tuning BERT for sentiment classification
• Fine tuning BERT variants for fake news detection
• Fine tuning DistilBERT variants for named entity recognition

Reference links

• Fine tuning Open AI documentation
• LLAMA 3 fine tuning with autotrain by hugging face
• LLAMA fine tuning on alpaca dataset using unsloth