Scaling Laws for Precision
This work addresses the cost-quality trade-off in low-precision ML for language models, offering a novel framework that is incremental in extending existing scaling laws.
The authors tackled the problem of how low precision training and inference affect language model quality and cost by developing precision-aware scaling laws that predict loss degradation from training and post-train quantization, validated on models up to 1.7B parameters and 26B tokens.
Low precision training and inference affect both the quality and cost of language models, but current scaling laws do not account for this. In this work, we devise "precision-aware" scaling laws for both training and inference. We propose that training in lower precision reduces the model's "effective parameter count," allowing us to predict the additional loss incurred from training in low precision and post-train quantization. For inference, we find that the degradation introduced by post-training quantization increases as models are trained on more data, eventually making additional pretraining data actively harmful. For training, our scaling laws allow us to predict the loss of a model with different parts in different precisions, and suggest that training larger models in lower precision may be compute optimal. We unify the scaling laws for post and pretraining quantization to arrive at a single functional form that predicts degradation from training and inference in varied precisions. We fit on over 465 pretraining runs and validate our predictions on model sizes up to 1.7B parameters trained on up to 26B tokens.