Scaling Laws
Scaling laws describe predictable relationships between model size, data, compute, and loss/performance. They guide how to allocate training budgets.
They are empirical trends with caveats — data quality, repetition, and architecture shifts can bend the curves.