Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
7
64
104
Aurélien-Morgan CLAUDON
Aurelien-Morgan
Follow
shayekh's profile picture
ltim's profile picture
umutphp's profile picture
9 followers
·
96 following
AurelienMorgan_
aurelienmorgan
aurelienmorganclaudon
aurelienmorgan.bsky.social
AI & ML interests
None yet
Recent Activity
liked
a Space
about 12 hours ago
akhaliq/anychat
replied
to
clem
's
post
1 day ago
Coming back to Paris Friday to open our new Hugging Face office! We're at capacity for the party but add your name in the waiting list as we're trying to privatize the passage du Caire for extra space for robots 🤖🦾🦿 https://t.co/enkFXjWndJ
reacted
to
m-ric
's
post
with 👍
1 day ago
𝐇𝐮𝐠𝐠𝐢𝐧𝐠 𝐅𝐚𝐜𝐞 𝐫𝐞𝐥𝐞𝐚𝐬𝐞𝐬 𝐏𝐢𝐜𝐨𝐭𝐫𝐨𝐧, 𝐚 𝐦𝐢𝐜𝐫𝐨𝐬𝐜𝐨𝐩𝐢𝐜 𝐥𝐢𝐛 𝐭𝐡𝐚𝐭 𝐬𝐨𝐥𝐯𝐞𝐬 𝐋𝐋𝐌 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝟒𝐃 𝐩𝐚𝐫𝐚𝐥𝐥𝐞𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧 🥳 🕰️ Llama-3.1-405B took 39 million GPU-hours to train, i.e. about 4.5 thousand years. 👴🏻 If they had needed all this time, we would have GPU stories from the time of Pharaoh 𓂀: "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 " 🛠️ 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. 🤏 𝗕𝘂𝘁 𝗻𝗼𝘄 𝘄𝗲 𝗱𝗼𝗻'𝘁 𝗻𝗲𝗲𝗱 𝗵𝘂𝗴𝗲 𝗿𝗲𝗽𝗼𝘀 𝗮𝗻𝘆𝗺𝗼𝗿𝗲! 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! ⚡ 𝗜𝘁'𝘀 𝘁𝗶𝗻𝘆, 𝘆𝗲𝘁 𝗽𝗼𝘄𝗲𝗿𝗳𝘂𝗹: 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 👉 https://github.com/huggingface/picotron/tree/main/picotron
View all activity
Articles
Fancy Stateful Metaflow Service + UI on Google Colab ?
Oct 14
•
4
Organizations
Aurelien-Morgan
's activity
All
Models
Datasets
Spaces
Papers
Collections
Community
Posts
Upvotes
Likes
New activity in
paris-ai-running-club/README
about 2 months ago
we're running a 10K: https://lagrandecourse.fr/
8
#4 opened 3 months ago by
julien-c
New activity in
open-llm-leaderboard/comparator
2 months ago
search selection list autocomplete
1
#1 opened 2 months ago by
Aurelien-Morgan
New activity in
blog-explorers/README
2 months ago
[Support] Community Articles
65
#5 opened 9 months ago by
victor
New activity in
compmicro-czb/VirtualStaining
3 months ago
Impressive
#1 opened 3 months ago by
Aurelien-Morgan
New activity in
paris-ai-running-club/README
5 months ago
next run wen?
8
#3 opened 5 months ago by
julien-c
New activity in
huggingface-projects/gemma-2-2b-it
5 months ago
Duplicate for private use
#1 opened 5 months ago by
Aurelien-Morgan
New activity in
huggingface/HuggingDiscussions
7 months ago
[FEEDBACK] Notifications
137
#6 opened over 2 years ago by
victor
[FEEDBACK] Notifications
137
#6 opened over 2 years ago by
victor