CVAIARApr 16, 2024

From a Lossless (~1.5:1) Compression Algorithm for Llama2 7B Weights to Variable Precision, Variable Range, Compressed Numeric Data Types for CNNs and LLMs

arXiv:2404.10896v11 citationsh-index: 1
Originality Incremental advance
AI Analysis

This addresses bandwidth reduction and storage efficiency for large models like CNNs and LLMs, though it appears incremental in extending existing compression techniques.

The paper introduces a lossless compression algorithm achieving ~1.5:1 compression for Llama2 7B weights, extendable to variable precision data types, with a hardware implementation processing over 800 million bfloat16 numbers per second.

This paper starts with a simple lossless ~1.5:1 compression algorithm for the weights of the Large Language Model (LLM) Llama2 7B [1] that can be implemented in ~200 LUTs in AMD FPGAs, processing over 800 million bfloat16 numbers per second. This framework is then extended to variable precision, variable range, compressed numerical data types that are a user defined super set of both floats and posits [2]. The paper then discusses a simple hardware implementation of such format based on ANS (Asymmetrical Numeral Systems) [3] that acts as a bridge between this flexible data format and a computational engine while, at the same time, achieving bandwidth reduction. An example of a token factory using weight compression and sharing is also given.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes