jonatasgrosman Profile Banner
Jonatas Grosman Profile
Jonatas Grosman

@jonatasgrosman

Followers
497
Following
158
Media
10
Statuses
80

PhD in Computer Science @pucriooficial

Joined October 2009
Don't wanna be here? Send us removal request.
@XDevelopers
Developers
3 months
Calling all developers! Innovate with our real-time and historical data on the X API. Get started with Pro
2K
6K
30K
@deepseek_ai
DeepSeek
9 months
🚀 Day 5 of #OpenSourceWeek: 3FS, Thruster for All DeepSeek Data Access Fire-Flyer File System (3FS) - a parallel file system that utilizes the full bandwidth of modern SSDs and RDMA networks. ⚡ 6.6 TiB/s aggregate read throughput in a 180-node cluster ⚡ 3.66 TiB/min
Tweet card summary image
github.com
A high-performance distributed file system designed to address the challenges of AI training and inference workloads. - GitHub - deepseek-ai/3FS: A high-performance distributed file system design...
529
1K
10K
@mvaloatto
Matt Valoatto
10 months
🤗 @HuggingFace’s most impactful contributors you should follow [January 2025] : https://t.co/93lSKxGgxk 🏛️ Top 10 Model downloads: 👏 @jonatasgrosman , @MaziyarPanahi , #Bingsu , #thenlper , @bartowski1182 , @intfloat , @ZhiyuQu_ , @very_laurie , @vikparuchuri , @LPapariello
3
16
62
@karpathy
Andrej Karpathy
1 year
The (true) story of development and inspiration behind the "attention" operator, the one in "Attention is All you Need" that introduced the Transformer. From personal email correspondence with the author @DBahdanau ~2 years ago, published here and now (with permission) following
136
991
7K
@mvaloatto
Matt Valoatto
1 year
🤗 @HuggingFace’s most impactful contributors you should follow [October 2024] : https://t.co/93lSKxGgxk - 🏛️ Top 10 Model downloads: 👏 @jonatasgrosman, #1231czx, @princeton_nlp, @intfloat, @prajjwal_1, @MaziyarPanahi, #mradermacher, @TheBlokeAI, @Emily_Alsentzer, #neuralmind
1
10
34
@mvaloatto
Matt Valoatto
1 year
🤗 @HuggingFace’s most impactful contributors you should follow [August 2024] : https://t.co/93lSKxGgxk 🏛️ Top 10 Model downloads: 👏 @jonatasgrosman, @patrickjohncyh, #avichr, @eugene_siow, @limsanity23, #MahmoudAshraf, @Emily_Alsentzer, @perezjotaeme, #intfloat, @TheBlokeAI
2
16
55
@hbredin
Hervé "pyannote" Bredin
1 year
Four (4 !) @pyannoteAI speaker diarization models among the 10 most downloaded audio models on @huggingface 🎉 (in good company with @jonatasgrosman @AIatMeta wav2vec2 and @OpenAI whisper)
0
1
32
@mvaloatto
Matt Valoatto
1 year
🤗 Here are the top 100 of @HuggingFace’s most impactful users of July 2024 (models, datasets, spaces, followers): 🏆 Top Contributors To Follow: https://t.co/93lSKxGgxk - 🏛️ Top 10 Model Downloads: 👏 @jonatasgrosman, #PatrickJohnChia, @Lin_Chen_98, @Emily_Alsentzer,
6
19
55
@mvaloatto
Matt Valoatto
1 year
🤗 Here are the top 100 of @HuggingFace’s most impactful users of June 2024 ( models, datasets, spaces, followers): 🏆 Top Contributors To Follow: https://t.co/93lSKxGgxk - 🏛️ Top 10 Model Downloads: 👏 @jonatasgrosman, #MichaelRadermacher, #avichr, @Kevin_E_Wu, @perezjotaeme,
2
8
29
@mvaloatto
Matt Valoatto
1 year
Happy Monday! 🤗 Here are the top 100 of @HuggingFace’s most impactful users of May 2024 ( models, datasets, spaces, followers): 🏆 Top Contributors To Follow: https://t.co/93lSKxGgxk 🏛️ Top 10 Model Downloads: 👏 @jonatasgrosman, @TheBlokeAI, @mrm8488, #LiangWang, #HosungPark,
2
16
64
@jonatasgrosman
Jonatas Grosman
1 year
I've just realized that my models had 86M downloads on HF in a single month 😮 ... thanks @mvaloatto for this space!
1
1
12
@ClementDelangue
clem 🤗
2 years
Top HF users by number of model downloads, dataset download, Spaces likes & profile followers. Congrats to everyone in there & to all who will get there soon! @mvaloatto created the space: https://t.co/TAi7SR8zER
3
9
38
@mvaloatto
Matt Valoatto
2 years
1
3
20
@mvaloatto
Matt Valoatto
2 years
Want more “good machine learning” in your X feed? Here is a new Space for you: 🔔 Top HF Users To Follow On X - https://t.co/jRaEF6hidq Ever since I fell down the AI rabbit hole, it hasn’t been super easy to spot and follow the most impactful @HuggingFace contributors on X. So,
Tweet card summary image
huggingface.co
6
23
86
@jeremyphoward
Jeremy Howard
2 years
Today, with @Tim_Dettmers, @huggingface, & @mobius_labs, we're releasing FSDP/QLoRA, a new project that lets you efficiently train very large (70b) models on a home computer with consumer gaming GPUs. 1/🧵 https://t.co/UAsWOLtn7a
Tweet card summary image
answer.ai
We’re releasing an open source system, based on FSDP and QLoRA, that can train a 70b model on two 24GB GPUs.
83
642
3K
@WIRED
WIRED
2 years
At the end of a year dominated by news of AI @StevenLevy sat down with @ylecun , pioneer of modern AI, chief AI scientist for Meta, and one of the technology's most vocal defenders, to ask why he believes AI danger is overhyped 📷 @eriktanner // 🔗 https://t.co/fAQNfGpY9T
13
71
230
@TimDarcet
TimDarcet
2 years
Vision transformers need registers! Or at least, it seems they 𝘸𝘢𝘯𝘵 some… ViTs have artifacts in attention maps. It’s due to the model using these patches as “registers”. Just add new tokens (“[reg]”): - no artifacts - interpretable attention maps 🦖 - improved performances!
40
310
2K
@TIIuae
Technology Innovation Institute
2 years
🌟 Introducing Falcon 180B: The World's Most Powerful Open LLM! 🚀 At #TII, we are continuing to push the boundaries of generative AI with our open access Falcon 180B AI model which has already soared to the top of the Hugging Face Leaderboard.
27
226
1K
@pucriooficial
PUC-RIO
2 years
Nosso ex-aluno, @jonatasgrosman, conquista destaque mundial com seu modelo ‘’Speech-To-Text’’, alcançando mais de 60 milhões de downloads e reconhecimento em vários idiomas. Link para download do modelo na matéria! https://t.co/7anJqgjiEj
Tweet card summary image
ctc.puc-rio.br
Modelo que transforma voz em texto, produzido por ex-aluno de Informática da PUC-Rio, já tem mais de 60 milhões de downloads
0
2
9
@ylecun
Yann LeCun
2 years
This is huge: Llama-v2 is open source, with a license that authorizes commercial use! This is going to change the landscape of the LLM market. Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers Pretrained and fine-tuned
394
3K
15K