March 13, 2024
Training Large Language Models (LLMs) presents significant memory challenges, predominantly due to the growing size of weights and optimizer states. Common memory-reduction approaches, such as low-rank adaptation (LoRA), add a trainable low-rank matrix to the frozen pre-trained weight in each layer, reducing trainable parameters and optimizer states. However, such approaches typically underperform training with full-rank weights in both pre-training and fine-tuning stages since they limit the parameter search to a low-rank subspace and alter the training dynamics, and further, may require full-rank warm start. In this work, we propose Gradient Low-Rank Projection (GaLore), a training strategy that allows full-parameter learning but is more memory-efficient than common low-rank adaptation methods such as LoRA. Our approach reduces memory usage by up to 65.5% in optimizer states while maintaining both efficiency and performance for pre-training on LLaMA 1B and 7B architectures with C4 dataset with up to 19.7B tokens, and on fine-tuning RoBERTa on GLUE tasks. Our 8-bit GaLore further reduces optimizer memory by up to 82.5% and total training memory by 63.3%, compared to a BF16 baseline. Notably, we demonstrate, for the first time, the feasibility of pre-training a 7B model on consumer GPUs with 24GB memory (e.g., NVIDIA RTX 4090) without model parallel, checkpointing, or offloading strategies.
Publisher
arXiv
Research Topics
Core Machine Learning
July 21, 2024
Ouail Kitouni, Niklas Nolte, Samuel Pérez Díaz, Sokratis Trifinopoulos, Mike Williams
July 21, 2024
July 08, 2024
Antonio Orvieto, Lin Xiao
July 08, 2024
June 17, 2024
Heli Ben-Hamu, Omri Puny, Itai Gat, Brian Karrer, Uriel Singer, Yaron Lipman
June 17, 2024
June 17, 2024
Neta Shaul, Uriel Singer, Ricky Chen, Matt Le, Ali Thabet, Albert Pumarola, Yaron Lipman
June 17, 2024
Product experiences
Foundational models
Product experiences
Latest news
Foundational models