Large Language Model Fine-tuning & Evaluation

10 repos

Tools and frameworks for adapting and evaluating large language models, particularly focused on instruction-following and specialized task performance. The cluster centers on open-source implementations of model fine-tuning pipelines (SCRIBE variants based on Llama models), evaluation frameworks for reasoning and long-context tasks, and educational resources for understanding model adaptation techniques. Projects here span both the practical engineering of model customization and the assessment methods needed to measure their effectiveness.

Related clusters