“Cramming: Training a Language Model on a Single GPU in One Day”, 2022-12-28 (; similar):
Recent trends in language modeling have focused on increasing performance through scaling, and have resulted in an environment where training language models is out of reach for most researchers and practitioners. While most in the community are asking how to push the limits of extreme computation, we ask the opposite question: How far can we get with a single GPU in just one day?
[code] We investigate the downstream performance achievable with a transformer-based language model trained completely from scratch with masked language modeling for 1 day on 1 consumer GPU [Nvidia RTX A6000]. Aside from re-analyzing nearly all components of the pretraining pipeline for this scenario and providing a modified pipeline with performance close to BERT, we investigate why scaling down is hard, and which modifications actually improve performance in this scenario.
We provide evidence that even in this constrained setting, performance closely follows scaling laws observed in large-compute settings… An unsurprising consequence of these laws is that scaling down is hard; while smaller model architectures enable speeding up gradient computations, overall rates of model improvement over time remain nearly constant. Nonetheless, we can find changes to the training recipe that exploit scaling laws to yield improvements by improving the effective rate of gradient computations without compromising model size.
Through the lens of scaling laws, we categorize a range of recent improvements to training and architecture and discuss their merit and practical applicability (or lack thereof) for the limited compute setting.