Open R1: Update #2

The Open R1 update introduces enhancements to the R1 model, including an increase in model size to 1.5 billion parameters and improvements in the training dataset, which now includes a broader range of domain-specific texts. Benchmark results indicate a 15% increase in performance on the GLUE tasks compared to the previous version. These updates are significant for practitioners as they provide a more robust model for natural language understanding tasks, enhancing the capabilities of applications built on the R1 architecture.