Hugging Face's logo
Soumik Rakshit's picture
Open to Collab
1 21 42

Soumik Rakshit

geekyrakshit
fmamberti-wandb's profile picture pcuenq's profile picture BenjaminDv's profile picture
ยท
http://geekyrakshit.dev

AI & ML interests

Computer vision

Recent Activity

reacted to m-ric's post with ๐Ÿ‘ about 11 hours ago
๐‡๐ฎ๐ ๐ ๐ข๐ง๐  ๐…๐š๐œ๐ž ๐ซ๐ž๐ฅ๐ž๐š๐ฌ๐ž๐ฌ ๐๐ข๐œ๐จ๐ญ๐ซ๐จ๐ง, ๐š ๐ฆ๐ข๐œ๐ซ๐จ๐ฌ๐œ๐จ๐ฉ๐ข๐œ ๐ฅ๐ข๐› ๐ญ๐ก๐š๐ญ ๐ฌ๐จ๐ฅ๐ฏ๐ž๐ฌ ๐‹๐‹๐Œ ๐ญ๐ซ๐š๐ข๐ง๐ข๐ง๐  ๐Ÿ’๐ƒ ๐ฉ๐š๐ซ๐š๐ฅ๐ฅ๐ž๐ฅ๐ข๐ณ๐š๐ญ๐ข๐จ๐ง ๐Ÿฅณ ๐Ÿ•ฐ๏ธ Llama-3.1-405B took 39 million GPU-hours to train, i.e. about 4.5 thousand years. ๐Ÿ‘ด๐Ÿป If they had needed all this time, we would have GPU stories from the time of Pharaoh ๐“‚€: "Alas, Lord of Two Lands, the shipment of counting-stones arriving from Cathay was lost to pirates, this shall delay the building of your computing temple by many moons " ๐Ÿ› ๏ธ But instead, they just parallelized the training on 24k H100s, which made it take just a few months. This required parallelizing across 4 dimensions: data, tensor, context, pipeline. And it is infamously hard to do, making for bloated code repos that hold together only by magic. ๐Ÿค ๐—•๐˜‚๐˜ ๐—ป๐—ผ๐˜„ ๐˜„๐—ฒ ๐—ฑ๐—ผ๐—ป'๐˜ ๐—ป๐—ฒ๐—ฒ๐—ฑ ๐—ต๐˜‚๐—ด๐—ฒ ๐—ฟ๐—ฒ๐—ฝ๐—ผ๐˜€ ๐—ฎ๐—ป๐˜†๐—บ๐—ผ๐—ฟ๐—ฒ! Instead of building mega-training codes, Hugging Face colleagues cooked in the other direction, towards tiny 4D parallelism libs. A team has built Nanotron, already widely used in industry. And now a team releases Picotron, a radical approach to code 4D Parallelism in just a few hundred lines of code, a real engineering prowess, making it much easier to understand what's actually happening! โšก ๐—œ๐˜'๐˜€ ๐˜๐—ถ๐—ป๐˜†, ๐˜†๐—ฒ๐˜ ๐—ฝ๐—ผ๐˜„๐—ฒ๐—ฟ๐—ณ๐˜‚๐—น: Counting in MFU (Model FLOPs Utilization, how much the model actually uses all the compute potential), this lib reaches ~50% on SmolLM-1.7B model with 8 H100 GPUs, which is really close to what huge libs would reach. (Caution: the team is leading further benchmarks to verify this) Go take a look ๐Ÿ‘‰ https://github.com/huggingface/picotron/tree/main/picotron
liked a dataset 1 day ago
ILSVRC/imagenet-1k
upvoted an article 15 days ago
Profiling in PyTorch (Part 1): A Beginner's Guide to torch.profiler
View all activity

Organizations

Restorers's profile picture TrystAI's profile picture Blog-explorers's profile picture Juspay Technologies's profile picture Ashwini AI's profile picture AthenaAgent's profile picture MalpuaAI's profile picture Brahmandam's profile picture