MMichał Piszczek·Nov 9Mode collapse is a default, not a law: what preference tuning does to varietyWhy 25 models give you two metaphors, and why temperature doesn't fix it11 min read·piszczek.hashnode.dev
TTechTIQ Inc·Oct 30RAG vs Fine-Tuning vs Prompt Engineering: Choosing the Right ApproachA practical guide to choosing between prompt engineering, RAG, and fine-tuning for your LLM project.4 min read·techtiqinc.hashnode.dev
RRebiha.com·Oct 20LLM Fine-Tuning Services: How to Choose the Right Provider in 2026A practical framework for evaluating model support, training data, QLoRA, privacy, evaluation, and model ownership.6 min read·rebiha.hashnode.dev
TTaha hussein·Oct 20Fine-Tuning vs LoRA vs QLoRA: Understanding Efficient LLM AdaptationLearn how Fine-Tuning, LoRA, QLoRA, and KD-LoRA help customize AI models. Explore key differences, benefits, and LLM adaptation techniques.5 min read·tahahussein.hashnode.dev
VViraj Sawant·Oct 15Directional Ablation Can't see 24° Rotation of the Refusal DirectionWhy directional ablation fails to detect refusal vector rotation after fine-tuning, and how calibration decouples estimate drift from safety.23 min read·mechanistic-artifacts.hashnode.dev
AAleksei Romanov·Sep 26LoRA & DoRA: The Math, Memory, and Trade-offsLoRA and DoRA explained: matrix shapes, the geometry of weight decomposition, exact memory math for a 27B model, and what the DoRA benchmarks support.10 min read·gfactor.hashnode.dev
FFotie M. Constant·Sep 23Project Turaco Is Finally Showing PromiseFor more than a year, I have been trying to answer one question: can we build a machine translation model that translates English into natural Cameroon Pidgin without sounding like it is guessing? Tha3 min read·blog.fotiecodes.com1B
TTaha hussein·Sep 18Fine-Tuning an LLM: What Actually Happens Under the Hood?Learn how LLM fine-tuning works under the hood, including training, parameters, LoRA, PEFT, and the differences between fine-tuning, RAG7 min read·tahahussein.hashnode.dev
RRahul Sai Indeevar V·Sep 16Parameter-Efficient Fine-Tuning: From LoRA to QLoRAFine-tuning Large Language Models (LLMs) used to be an option only for organizations with massive compute budgets. Full fine-tuning requires updating and storing every parameter, gradient, and optimiz10 min read·rahul-ai.hashnode.dev2
RRahul Sai Indeevar V·Sep 8LLM Post-Training: From Full Fine-Tuning to PEFT, Adapters, and Soft PromptsLarge Language Models (LLMs) arrive with immense foundational knowledge acquired during pre-training. However, deploying them to specialized real-world domains—such as legal document parsing, medical14 min read·rahul-ai.hashnode.dev1