Fine-tuning a LLM on my blog posts | Didier Lopes

At the start of the year I shared this blogpost, which converted all my blogs into a Q&A dataset that I could use to fine-tune a LLM. After sharing this, I spent time trying to fine-tune an LLM - but results were not great and combined with all things happening at OpenBB I didn't have time to dedicate a lot of time to this. :upside-down-hf-logo But I hate leaving things half way. And this task didn’t leave my TODO for the past 6 months. So I finally took things to my hands last weekend, and I’m going to share the entire journey on what, why and how. Buckle up, this will be a long post - and more technical than previous ones. And all the code will be available here: https://github.com/DidierRLopes/fine-tune-llm. Most AI models are like Wikipedia - they know a little about everything but lack the depth and personality that comes from lived experience. Think of it this way: RAG is like giving someone a reference book during an exam. Fine-tuning is like actually teaching them the subject u

Ever wondered what it would be like to have an AI that writes exactly in your style? I did. And in this post, I share what I did about it. This is a very practical guide on how to fine-tune an LLM using LoRA with MLX on Apple Silicon. At the start of the year I shared this blogpost , which converted all my blogs into a Q&A dataset that I could use to fine-tune a LLM. After sharing this, I spent time trying to fine-tune an LLM - but results were not great and combined with all things happening at OpenBB I didn't have time to dedicate a lot of time to this. :upside-down-hf-logo But I hate leavin

related reading

• Fine-Tuning Llama-2: Tailoring Models to Unique Applicationsanyscale.com
• LoRA Without Regret - Thinking Machines Labthinkingmachines.ai
• Patterns for Building LLM-based Systems & Productseugeneyan.com
• Fine-tuning LLMs Guide | Unsloth Documentationdocs.unsloth.ai
• The Llama Hitchiking Guide to Local LLMs – hackerllamaosanseviero.github.io
• Finetune LLMs on your own consumer hardware using tools from PyTorch and Hugging Face ecosystem – PyTorchpytorch.org
• The Ultimate Guide to Fine-Tuning LLMs from Basics to Breakthroughs: An Exhaustive Review of Technologies, Research, Best Practices, Applied Research Challenges and Opportunities (Version 1.0)arxiv.org
• Tutorial: How to Finetune Llama-3 and Use In Ollama | Unsloth Documentationdocs.unsloth.ai
• Model optimization | OpenAI APIplatform.openai.com
• Efficient LLM Finetuning with Unsloth | Modal Docsmodal.com
• GitHub - bobazooba/xllm: 🦖 X—LLM: Cutting Edge & Easy LLM Finetuning · GitHubgithub.com
• Parameter-Efficient LLM Finetuning With Low-Rank Adaptation (LoRA) - Lightning AIlightning.ai