Training Tiny LLMs to Play Connections Better than Frontier Models
header
Brian Fitzgerald Blog
Brian Fitzgerald Blog
Latest posts
Training a Reasoning Model for Very Cheap with DPO
header
Reviewing Post-Training Techniques from Recent Open LLMs
header
Curriculum Learning with DPO and Logit Pruning
Background
Tiny Agents - Training Small LLMs to Use Tools with DPO and Synthetic Data
TL;DR: I've created a synthetic dataset for training LLMs to use tools, and am training a model using DPO to improve accuracy. The dataset is available here, and I'll be releasing trained models soon.
SuperPrompt - Better SDXL prompts in 77M Parameters
Left SDXL output with SuperPrompt applied to the same input prompt.
Understanding StyleAligned
For the last two years, the improvement in fidelity and control in diffusion models has been incredible; generation results are often indistinguishable from real images, and a huge ecoystem of tools has emerged to offer various forms of control over the generation process. However, for a lot of real-world workflows, you quickly run up against limitations that make things hard for less technical…
Longer videos with Stable Video Diffusion via YaRN
Background