Joonhyung Lee's picture

2

Joonhyung Lee

joonhyung-lee-naver

·

AI & ML interests

None yet

Recent Activity

liked a Space 4 days ago

nanotron/ultrascale-playbook

reacted to regisss's post with 🔥 8 days ago

Nice paper comparing the fp8 inference efficiency of Nvidia H100 and Intel Gaudi2: https://huggingface.co/papers/2502.01070 The conclusion is interesting: "Our findings highlight that the Gaudi 2, by leveraging FP8, achieves higher throughput-to-power efficiency during LLM inference" One aspect of AI hardware accelerators that is often overlooked is how they consume less energy than GPUs. It's nice to see researchers starting carrying out experiments to measure this! Gaudi3 results soon...

authored a paper 7 months ago

HyperCLOVA X Technical Report

View all activity

Organizations

None yet

Papers 2

arxiv:2405.18710

arxiv:2404.01954

models

None public yet

datasets

None public yet