Recipients of Uplimit’s Finetuning Open Source LLMs certificate have gained foundational skills in fine-tuning large language models (LLMs) for custom and domain-specific applications. The course, designed for data scientists, machine learning engineers, and AI researchers, covered advanced techniques like Direct Preference Optimization (DPO), Supervised Fine-Tuning (SFT), and Orthogonal Random Parameter Optimization (ORPO), along with data-centric evaluation, benchmarking, and model quantization for productizing models.

Certification criteria:

Actively participated in Uplimit’s dynamic course on Finetuning Open Source LLMs: attended lectures, discussion sessions, and more.

Built a customized, fine-tuned large language model (LLM) optimized for specific applications through evaluation, fine-tuning, and deployment techniques, ready for production use.

November 4, 2024 Instructor: Ben Burtenshaw

issued by

Learners gained practical skills in selecting evaluation metrics, preparing high-quality datasets, and deploying models efficiently, empowering them to adapt LLMs to custom, domain-specific needs.