4 accepted papers
An increasing number of LLMs employ Mixture-of-Experts (MoE) architectures where the feed-forward layer is replaced by a pool of experts and each token only activates a small subset of them. During autoregressive generation, these models often enter a memory-bound regime even for moderate batch size…
Increasing the batch size during training --- a “batch ramp'' --- is a promising strategy to accelerate large language model pretraining. While for SGD, doubling the batch size can be equivalent to halving the learning rate, the optimal strategy for adaptive optimizers like Adam is less clear. As a…
While transformers have been at the core of most recent advancements in sequence generative models, their computational cost remains quadratic in sequence length. Several subquadratic architectures have been proposed to address this computational issue. Some of them, including long convolution seque…
Understanding the internal representations learned by neural networks is a cornerstone challenge in the science of machine learning. While there have been significant recent strides in some cases towards understanding *how* neural networks implement specific target functions, this paper explores a c…