How to Fine-Tune a LLM on Your Own Data

In these guides, we run a LLM, and then use QLoRA (Quantized Low-Rank Adaptation) to cost-effectively fine-tune it on our own dataset.

I fine-tuned the model on my electronic journal entries from over the years, and it was honestly hilarious to see it write like an angsty teenage Harper.

Watch the tutorial on YouTube:

Watch the Finetuning Tutorial

Jupyter Notebooks

Click here to read about how QLoRA works to reduce computational complexity when finetuning!