SpikeLLM: Scaling up Spiking Neural Network to Large Language Models via Saliency-based Spiking
Keywords: spiking neural network, energy efficient language model, brain-inspired computing
Abstract: The recent advancements in large language models (LLMs) with billions of parameters have significantly boosted their performance across various real-world applications. However, the inference processes for these models require substantial energy and computational resources, presenting considerable deployment challenges. In contrast, human brains, which contain approximately 86 billion biological neurons, exhibit significantly greater energy efficiency compared to LLMs with a similar number of parameters. Inspired by this, we redesign 7~70 billion parameter LLMs using bio-plausible spiking mechanisms, emulating the efficient behavior of the human brain. We propose the first spiking large language model termed SpikeLLM. Coupled with the proposed model, two essential approaches are proposed to improve spiking efficiency: Generalized Integrate-and-Fire (GIF) neurons to compress spike length from $T$ to $\frac{T}{L} \log_2 L$ bits, and an Optimal Brain Spiking framework to divide outlier channels and allocate different $T$ for GIF neurons, which further compresses spike length to approximate $log_2T$ bits. The necessity of spike-driven LLM is proved by comparison with quantized LLMs with similar operations. In the OmniQuant pipeline, SpikeLLM reduces 24.85% WikiText2 perplexity and improves 2.01% accuracy of common scene reasoning on a LLAMA2-7B 4A4W model. In the GPTQ pipeline, SpikeLLM achieves direct additive in linear layers, significantly exceeding PB-LLMs. In the LLAMA-2-7B, SpikeLLM saves $\times 10.79$ and $\times 6.38$ operations with general matrix multiply and event-driven implementations respectively. We will release our code on GitHub.
Primary Area: foundation or frontier models, including LLMs
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 7612
Loading