ML & AI 5
All writing- Teaching MiniCPM5 1B and 2B to be SQL agents on a MacBook
Training small models to answer database questions with SQL tools on one MacBook. On question types they never trained on, MiniCPM5-1B goes from 3% to 48% and the 2B from 38% to 73%.
- LLM fundamentals: pretraining, fine-tuning and evaluation
How language models generate text, how training methods differ, and how to interpret results from LoRA, QLoRA, LIMA, DPO and reasoning-model research.
- RAG: diagnosing retrieval and answer failures
A practical guide to checking ingestion, retrieval, context, citations and permissions in a retrieval-augmented generation system.
- Running LLMs on a Steam Deck in 5 minutes
A step-by-step guide to setting up your Steam Deck to run local Large Language Models (LLMs) using llama.cpp and a Distrobox container.
- Fine-tuning a 3B Model for Transcript Analysis: Better Than 70B Models?
How I fine-tuned Llama 3.2 3B to process transcripts locally, achieving better results than models 5-8x larger through specialized training.