English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

FP8 formats are gaining popularity to boost the computational efficiency fortraining and inference of large deep learning models. Their main challenge isthat a careful choice of scaling is needed to prevent degradation due to thereduced dynamic range compared to higher-precision formats. Although thereexists ample literature about selecting such scalings for INT formats, thiscritical aspect has yet to be addressed for FP8. This paper presents amethodology to select the scalings for FP8 linear layers, based on dynamicallyupdating per-tensor scales for the weights, gradients and activations. We applythis methodology to train and validate large language models of the type of GPTand Llama 2 using FP8, for model sizes ranging from 111M to 70B. To facilitatethe understanding of the FP8 dynamics, our results are accompanied by plots ofthe per-tensor scale distribution for weights, activations and gradients duringboth training and inference.

Training and inference of large language models using 8-bit floating point