Absence of a Universal Scaling Law

A research team is developing a new language model. They train several versions of the model, each with a different number of parameters, while keeping the training dataset size fixed. They plot the final training loss for each model version against its parameter count. The resulting graph shows a consistent, downward-curving trend: as the number of parameters increases, the loss decreases, but the amount of improvement gets smaller with each increase. Based on this observation, what is the most accurate conclusion the team can draw?

Optimizing LLM Training Budget

Comparing Single-Variable Scaling Functions