Fine tuning

Fine-tuning is necessary when you want to adjust the style, tone, or format. 
It is used to improve reliability and accuracy, handle complex prompts, 
or perform a new task that the prompt engineer could not achieve.

Why Fine-Tune Language Models?

• Customization for specific use cases
• Improved steerability and reliability
• Enhanced performance

List of demo

• Working with OpenAI-gpt-4o
• Sentiment analysis with GPT2
• Mistral 7B 4bit QLoRA Fine-tuning

Reference websites

• OpenAI Fine-tuning
• Fine-tuning now available for GPT-4o
• https://huggingface.co/learn/llm-course/en/chapter11/1
• https://docs.opentyphoon.ai/th/fine-tuning-deployment/
• https://huggingface.co/docs/trl/en/sft_trainer