Abstract:
Large Transformer models yield impressive results on many tasks, but are expensive to train, or even fine-tune, and so slow at decoding that their use and study becomes out of reach. We address this problem by leveraging sparsity. We study sparse variants for all layers in the Transformer and propose Scaling Transformers, a family of next generation Transformer models that use sparse layers to scale efficiently and perform unbatched decoding much faster than the standard Transformer as we scale up the model size. Surprisingly - the sparse layers are enough to obtain the same perplexity as the standard Transformer. We also integrate with prior sparsity approaches to enable fast inference on long sequences even with limited memory, resulting in performance competitive to the state-of-the-art on long text summarization.
