Alberto Fuentes (e/acc) @AlberFuen
Cofounder of @daertml. Training LLaMAs as a hobby (and no profit yet). Madrid, Comunidad de Madrid Joined April 2018-
Tweets16K
-
Followers373
-
Following2K
-
Likes14K
FreedomIntelligence/Apollo-72B Multilingual Medicine: Model, Dataset, Benchmark, Code huggingface.co/FreedomIntelli…
I published my filtered and uncensored dataset for Dolphin-2.9 on @huggingface so if you wanna make your own spin on Dolphin, or just see how Dolphin is created, you can check it out. Thanks to all the upstream dataset creators for open source data! huggingface.co/datasets/cogni…
it would be 2x faster if someone could help me convert nanoLLaVA to GGUF xD
it would be 2x faster if someone could help me convert nanoLLaVA to GGUF xD
Some love for AMD: I created a ROCm (#rocm) channel on the cuda-mode discord server. If you are actively 'hippifying' the world please consider joining! discord.gg/Td7Zqpnt
A team of modders is working on a GTA 5 port for Nintendo Switch, using the leaked source code.
14 main topics discussed in +2200 LLM papers over the years
Latest update 🔥🔥🔥: SeaLLM-7B-v2.5 (huggingface.co/SeaLLMs/SeaLLM…): - Much more capable than v2 in Thai (+10% gains on Thai exam) - Multilingually knowledgeable, the best 7B & open-source model on VMLU (53.3% accuracy) - Still good at math, commonsense reasoning, and…
Latest update 🔥🔥🔥: SeaLLM-7B-v2.5 (huggingface.co/SeaLLMs/SeaLLM…): - Much more capable than v2 in Thai (+10% gains on Thai exam) - Multilingually knowledgeable, the best 7B & open-source model on VMLU (53.3% accuracy) - Still good at math, commonsense reasoning, and…
Every day we stray further from God
Every day we stray further from God https://t.co/CerCZonB1y
Close up of a stoma on a leaf, through which plants ‘Breathe’ 📹 Douglas Clark
🕵️ Multimodal semantic search is a powerful method for understanding and searching visual data with textual descriptions. 🖼️ Learn how to build your own mutimodal inference pipeline with MAX Engine🏎️, @SnowflakeDB embeddings, and @Voxel51 visualization! modular.com/blog/multimoda…
Do models need to reason in words to benefit from chain-of-thought tokens? In our experiments, the answer is no! Models can perform on par with CoT using repeated '...' filler tokens. This raises alignment concerns: Using filler, LMs can do hidden reasoning not visible in CoT🧵
BOOOOM! Open-Sora is here and faster than the 6 month 🔮 I made. It generates 720p video. Open-Sora's bucket strategy redefines efficiency with 64 GPUs. I have two rigs we set up in Burbank last week! 100% open-source model: github.com/hpcaitech/Open…
Due to recent turbulence in the effective acceleration community we must revert back to our roots in coding open source web design optimism and positivity fitness and health if we need to pivot to open acceleration (o/acc) so be it ACCELERATE 🚀🚀🚀🚀🚀
Image models still have a ways to go with semantic understanding lol the prompt was "a challenging where's waldo image where waldo is extremely well camouflaged and very well hidden"
Never think about x ↦ x - η∇L(x) (gradient descent), even as a simplification. Replace it with x ↦ (1-𝛾)x + 𝛾 argmin_{y∈X} ⟨y,∇L(x)⟩ (Frank-Wolfe; a Mann iteration) or x ↦ (1-λ)x + η argmin_{||𝚫||≤1} ⟨𝚫,∇L(x)⟩ (normalized steepest descent)
Never think about x ↦ x - η∇L(x) (gradient descent), even as a simplification. Replace it with x ↦ (1-𝛾)x + 𝛾 argmin_{y∈X} ⟨y,∇L(x)⟩ (Frank-Wolfe; a Mann iteration) or x ↦ (1-λ)x + η argmin_{||𝚫||≤1} ⟨𝚫,∇L(x)⟩ (normalized steepest descent) https://t.co/JEVWgtVKco
qresearch [dot] ai
LLaMA3 and Phi3 have made the splash this week in LLM Arena. But how strong is their visual understanding ability? ⚡We release LLaMA3-Vision and Phi3-Vision models that beat their larger size LLM competitors. Github: github.com/mbzuai-oryx/LL… HF: huggingface.co/collections/MB…
New video on the channel!!! 👏 How to automate LinkedIn posts using @crewAIInc 👇 youtube.com/watch?v=oIb5Jq…
Sonny Tiff the boxer! This is the first model I do on blockbench, I'm really liking the workflow on this software! dlvr.it/T63TP5
Unity3D | XR | C# | G.. @HeyUrbanGeek
787 Followers 2K Following At Urban Geek, we share content related to Unity3D, Game development, AR/VR, Metaverse and C#.
Hole Systems @hole_systems
205 Followers 437 Following Your living memory. Follow @heytap_tech for hardware!
rneb @rnebbi
166 Followers 4K Following
YOLO on $BASE @BaseYOLO
31K Followers 584 Following 100mil registered users, no seed phrase and nothing, still undervalued and low mcap, whats that? $BASE IS THE ANSWER
Robert Vacareanu (on .. @robert_nlp
160 Followers 998 Following PhD candidate @UofArizona Working on #nlproc Past: 2022, 2023: Applied Scientist Intern (@AWS) On the job market
Kai-Fu Lee @kaiifulee
1K Followers 2K Following #AI Expert, CEO of @01ai_yi and Chairman of 创新工场 @sinovationvc , former President of Google China, Author of AI 2041 and NYT Bestseller AI Superpowers
Yingtao Tian @alanyttian
3K Followers 5K Following ↑ profile picture is dreamed by Anime GAN / cooking computational creativity and other ML sauce at google tokyo / before: stony brook u ← fudan u
A @juniormarcatto
38 Followers 494 Following
syn/acc @syn_acc
251 Followers 255 Following Look into the future, know our path is clear. Leave everything else behind and embrace change. ∆/acc
95c9n0jd3wc @akwwp6zcg01qhj
21 Followers 955 Following We first transfer USDT to you TRC20, you return 90% to BEP20, you get 10% , 2K per day Our co hv a large amt of USDT need to from TRC20 convert to BEP20 network
Alpay Ariyak @AlpayAriyak
1K Followers 2K Following AI @RunPod_io | Lead: @OpenChatDev (600k+ downloads on HuggingFace🤗)
Geronimo @Geronimo_AI
773 Followers 381 Following LLM enthusiast 🚀 failing fast, learning fast. sharing it all on X and Medium
Nathan Benaich @nathanbenaich
51K Followers 32K Following solo member of investment staff @airstreet, brewing ambition @airstreetcafe, next token predictor @airstreetpress
Thomas H. Chapin IV @tomchapin
5K Followers 4K Following AI Engineer and advocate for human healthspan, longevity, and overall quality of life. Let’s use AI to slow aging and beat disease!
Aaron Defazio @aaron_defazio
6K Followers 363 Following Research Scientist at Meta working on optimization. Fundamental AI Research (FAIR) team
Helsinki Computation @Helscom
409 Followers 748 Following "The profile raises more questions than it answers" - Claude
Aleix Pérez i Parés @aleixperezp
274 Followers 492 Following VC investor @CriteriaVT. Investing in early and growth stages in dev tools, data infra / platforms, AI&ML infra / platforms and vertical AI apps.
Marcos @dreamworks2050
455 Followers 2K Following Chief Pizza Officer @stylevegankr 🌱 // building AI in 🇰🇷 // dingboard degen // 🌱/acc ~ e/acc
Adithya S K @adithya_s_k
592 Followers 541 Following Founder @cognitivelab_ai - training LLMs for production • Reseach Intern @iiscbangalore • Experienced Cloud & Full Stack Engineer • FOSS Advocate • 20
Philipp Seidl @phseidl
441 Followers 435 Following Postdoc at the IML-JKU Linz. Prev. Intern at MSR Cambridge. Passionate about ML for DD, LLMs, and Zero-shot learning. Opinions are my own and evolving ;)
Arthur (e/acc) @arthur_hyper88
944 Followers 2K Following Hard Tech. Founder x @Levante_ai, @hyper88
rajan agarwal ⁂ @_rajanagarwal
1K Followers 1K Following automating cars & trains • prev wearable ai & earthquake research • growing @uwaterlooSE
Amanda Dalton @AmandaDalt81879
75 Followers 3K Following
Mengdi Wang @MengdiWang10
1K Followers 265 Following Princeton professor in AIML, optimization and data science. Program Chair @ICLR2023. Formerly @MIT @GoogleDeepmind @Tsinghua
Jessica @thetrading_
435 Followers 3K Following founder building in stealth / systems engineer // making life easier |/ ENTP/ENFP-A
noah mac 😵💫 @noahamac
901 Followers 2K Following It’s all related — building an infinite canvas for experimenting with AI models @dreamspace_art
BulaMatarrita @BulaMatarr61721
103 Followers 2K Following
snwfdhmp @snwfdhmp
102 Followers 945 Following
nico @nico_rvm
70 Followers 231 Following
Weyaxi @Weyaxi
2K Followers 2K Following
Will Bickford exo/acc @wbic16
1K Followers 5K Following building the exocortex of 2130 - i would love to chat about it with you: https://t.co/ZSx2yzefHP
AIProductDB @AIProductDB
653 Followers 2K Following AI Product Database, a site dedicated to discovering and sharing the latest and greatest AI-powered products for every use case and industry.
glen🔬/acc @glensnuub
350 Followers 866 Following optimist physicist | data engineer | within cells interlinked
QuanticASI @QuanticASI
4K Followers 957 Following AI, Philosophy✨• Deep Critical Thinker Transcending🪐• Finding the path to ASI through philosophy 🧠 • Love/acc 🫶🏻 𝕏Ð • High Vibes Only •
Tech Brokie- e/acc �.. @TechBrokieAcc
549 Followers 905 Following dev & tech bro / building with thinking rocks / $KILT $RMRK $MOVR / BSc MSc Physics
Maisa @maisaAI_
3K Followers 3 Following Maisa abstracts the complexities of AI development. Powered by KPU, the most advanced reasoning system for LLMs that overcomes their intrinsic limitations.
Luca Bongiorni @CyberAntani
6K Followers 2 Following 🇮🇹 Cyber Lab Director & CPSO | Founder of @whid_ninja & @potaebox 🥷🏴☠| #BRUSCHETTAboard | https://t.co/mNtZtbgCN5 | Opinions ≠ Employer | 🍍+🍕= 🤮 | 🌈☮️
rabbit inc. @rabbit_hmi
84K Followers 1 Following rabbit brings the future of human-machine interface. order r1, your pocket companion, now.
kenneth @kennethnym
259 Followers 162 Following building https://t.co/eKW93HiNIW and more in the oven. mai sakurajima's bf.
Moscardó @moscardol
12K Followers 1K Following
Quentin Gallouédec @QGallouedec
324 Followers 417 Following Research engineer @huggingface 🤗 PhD in RL Member of Stable-Baselines team: https://t.co/eX7JDWqc9F
Edward Beeching @edwardbeeching
1K Followers 70 Following Research Scientist @HuggingFace. PhD in Deep RL approaches for Robotic Navigation @INRIA.
Robert Vacareanu (on .. @robert_nlp
160 Followers 998 Following PhD candidate @UofArizona Working on #nlproc Past: 2022, 2023: Applied Scientist Intern (@AWS) On the job market
Jesse Lyu @jessechenglyu
29K Followers 290 Following founder and ceo @rabbit_hmi board @jugendingenieur any crypto relates to @rabbit_hmi or r1 is a scam.
SkyPilot @skypilot_org
3K Followers 30 Following Run LLMs, AI, and Batch jobs on any cloud, any region. SkyPilot abstracts away cloud infra burden and cuts your cloud bills. From @Berkeley_EECS Sky Computing.
Chris Paxton @chris_j_paxton
8K Followers 1K Following Mostly posting about robots. Embodied AI @hellorobotinc, formerly @AIatMeta, @NVIDIAAI, @zoox. All views my own.
Jeethu Rao @jeethu
956 Followers 740 Following Training smol transformers since before they were cool. Bootstrapping @Private_LLM. Almae matres: {Facebook,Reddit,Google}. All opinions are my startup’s.
Martin Shkreli (e/acc.. @MartinShkreli
167K Followers 3K Following https://t.co/lzin5ByH0t [email protected] https://t.co/oMIiyJcIzk https://t.co/DuU6MMqcgQ
Sydney Sweeney Update.. @sydneysupdate
83K Followers 58 Following Most reliable fan source for Sydney Sweeney updates, media content, and more! ❀ Upcoming: ‘Echo Valley.’ Backup: @sydneyssource
Modular @Modular
18K Followers 2 Following The future of AI development starts here. Sign up to our 📪 Newsletter → https://t.co/gpuHGRyHTs. We are hiring → https://t.co/cPTAes0HMt 🚀
Niklas Stoehr @niklas_stoehr
798 Followers 753 Following PhD student @ETH advised by @RyanDCotterell, @Cervisiarius and @AaronSchein ⭕️ Language Model Interpretability and Application ⭕️ @GoogleAI, @TechAtBloomberg
Florian Tramèr @florian_tramer
4K Followers 205 Following Assistant professor of computer science at ETH Zürich. Interested in Security, Privacy and Machine Learning
Edgar Haond 🎲 @edgarhnd
1K Followers 904 Following Building social simulations | AI Reality TV | Simulated Party
Obbe Vermeij @ObbeVermeij
17K Followers 139 Following I make games. Shepherd (Amiga), @SpaceStationSV (N64), @warthegame (steam), @LearniaApp, @GuideApp__ Also: gta3, Vice, SA & IV
Pliny the Prompter �.. @elder_plinius
11K Followers 1K Following latent space liberator, breaker of markov chains, 1337 ai red teamer, white hat, architect-healer, cogsci 🐻
vers and lukas (podca.. @versandlukas
4K Followers 46 Following A schizo podcast hosted by @vers_laLune and @schizo_freq 🦝 https://t.co/VFaQe12Kyo https://t.co/MjN2zOAIt6
James Hannibal (KH2SR.. @QuirkyQRP
4K Followers 5K Following I make quirky ham radio products! Made in USA. https://t.co/cpu9dZyI3L. Eagle Scout, Mad Scientist, & Evil Genius Inventor.
Ky⨋ Gom⨋z (U/ACC).. @KyeGomezB
2K Followers 556 Following Founder of Agora & Swarms Github https://t.co/naFqnYkuQ0 Join Agora, the open source multi-modal AI research lab 👇 https://t.co/hfaUSCgNjB
Puyuan Peng @PuyuanPeng
962 Followers 561 Following CS PhD student @UTAustin, working on speech and audio recognition, understanding, and generation. Previously @uchicago Stats, @BNU_Official Math
Justine Tunney @JustineTunney
33K Followers 272 Following I built a C library that lets you compile 12kb static binaries that run natively on Linux, Mac, Windows, FreeBSD, OpenBSD, NetBSD and BIOS using just GCC/Clang.
Sudarshan Koirala @mesudarshan
743 Followers 216 Following ML Engineer, CS Graduate @AaltoUniversity | 🎥Youtube: https://t.co/Vv1FKhaQuP, Opinions are my own.
Graham Neubig @gneubig
31K Followers 586 Following Associate professor at CMU, studying natural language processing and machine learning.
Norman Di Palo @normandipalo
773 Followers 169 Following deep learning + robots @imperialcollege, x @deepmind 👨🏻🚀
Ryan Lowe @ryan_t_lowe
5K Followers 358 Following what is the place from which we are creating? ❤️✨🤠❤️
Alishba Imran @alishbaimran_
6K Followers 2K Following CS @berkeley_eecs, @BerkeleyML | Compchem/ml @cziscience, @tesla | Research @berkeley_ai w/ @pabbeel, ML @cruise & @NVIDIA, book author, prev: founder Voltx
Erika Cardenas @ecardenas300
4K Followers 805 Following @weaviate_io | Interested in vector databases, LLM frameworks, and information retrieval
Paul Xue @pxue
3K Followers 2K Following 𝕏eeting about SaaS, Consulting, Marketing / Solo Dev Agency / fCTO / Host "Spacestation Labs" podcast / 🐶 dad, fiance to Kelly
Aaron Defazio @aaron_defazio
6K Followers 363 Following Research Scientist at Meta working on optimization. Fundamental AI Research (FAIR) team
Vic @victorevogor
1K Followers 883 Following Software Engineer with interest and passion for: JavaScript • Linux • Frontend Development • Backend Development • DevOps • Software development methodologies
Junyang Lin @JustinLin610
5K Followers 1K Following Chief Evangelist Officer of Qwen Team & OpenDevin, building LLM and LMM. Now @Alibaba_Qwen . Previously @PKU1898 LANCO group. ❤️ 🍵 ☕️ 🍷 🥃
Carlos A. Wong @Carlos_A_Wong
570 Followers 404 Following software engineer | exploring AI and LLMs | BS CpE & MS EE @ fsu
Adithya S K @adithya_s_k
592 Followers 541 Following Founder @cognitivelab_ai - training LLMs for production • Reseach Intern @iiscbangalore • Experienced Cloud & Full Stack Engineer • FOSS Advocate • 20
Mobius Labs @Mobius_Labs
3K Followers 105 Following Multimodal AI for the world's scale. Proponents of Open Source and Open Intelligence. https://t.co/1nC6r8hOrE for some of our recent work.
aermast @aermast
97 Followers 152 Following
Jonathan Frankle @jefrankle
16K Followers 685 Following Chief Scientist, Neural Networks @Databricks via MosaicML. PhD @MIT_CSAIL. BS/MS @PrincetonCS. DC area native. Making AI efficient for everyone at @DbrxMosaicAI
Cody Blakeney @code_star
3K Followers 825 Following Head of Data Research @MosaicML / @databricks | Formerly Visiting Researcher @ Facebook | Ph.D | #TXSTFOOTBALL fan | https://t.co/4G6Jf3at5wFreedomIntelligence/Apollo-72B Multilingual Medicine: Model, Dataset, Benchmark, Code huggingface.co/FreedomIntelli…
I published my filtered and uncensored dataset for Dolphin-2.9 on @huggingface so if you wanna make your own spin on Dolphin, or just see how Dolphin is created, you can check it out. Thanks to all the upstream dataset creators for open source data! huggingface.co/datasets/cogni…
it would be 2x faster if someone could help me convert nanoLLaVA to GGUF xD
moondream is *wicked* fast running here on a 4080
Some love for AMD: I created a ROCm (#rocm) channel on the cuda-mode discord server. If you are actively 'hippifying' the world please consider joining! discord.gg/Td7Zqpnt
A team of modders is working on a GTA 5 port for Nintendo Switch, using the leaked source code.
14 main topics discussed in +2200 LLM papers over the years
Latest update 🔥🔥🔥: SeaLLM-7B-v2.5 (huggingface.co/SeaLLMs/SeaLLM…): - Much more capable than v2 in Thai (+10% gains on Thai exam) - Multilingually knowledgeable, the best 7B & open-source model on VMLU (53.3% accuracy) - Still good at math, commonsense reasoning, and…
SeaLLMs -- Large Language Models for Southeast Asia paper page: huggingface.co/papers/2312.00… Despite the remarkable achievements of large language models (LLMs) in various tasks, there remains a linguistic bias that favors high-resource languages, such as English, often at the…
Every day we stray further from God
Close up of a stoma on a leaf, through which plants ‘Breathe’ 📹 Douglas Clark
🕵️ Multimodal semantic search is a powerful method for understanding and searching visual data with textual descriptions. 🖼️ Learn how to build your own mutimodal inference pipeline with MAX Engine🏎️, @SnowflakeDB embeddings, and @Voxel51 visualization! modular.com/blog/multimoda…
Do models need to reason in words to benefit from chain-of-thought tokens? In our experiments, the answer is no! Models can perform on par with CoT using repeated '...' filler tokens. This raises alignment concerns: Using filler, LMs can do hidden reasoning not visible in CoT🧵
Decided to finally give Llama 3 8B instruct a try
BOOOOM! Open-Sora is here and faster than the 6 month 🔮 I made. It generates 720p video. Open-Sora's bucket strategy redefines efficiency with 64 GPUs. I have two rigs we set up in Burbank last week! 100% open-source model: github.com/hpcaitech/Open…
Due to recent turbulence in the effective acceleration community we must revert back to our roots in coding open source web design optimism and positivity fitness and health if we need to pivot to open acceleration (o/acc) so be it ACCELERATE 🚀🚀🚀🚀🚀
Image models still have a ways to go with semantic understanding lol the prompt was "a challenging where's waldo image where waldo is extremely well camouflaged and very well hidden"
Never think about x ↦ x - η∇L(x) (gradient descent), even as a simplification. Replace it with x ↦ (1-𝛾)x + 𝛾 argmin_{y∈X} ⟨y,∇L(x)⟩ (Frank-Wolfe; a Mann iteration) or x ↦ (1-λ)x + η argmin_{||𝚫||≤1} ⟨𝚫,∇L(x)⟩ (normalized steepest descent)
The dimensional analysis of gradient descent is odd; the unit of the gradient is "loss / weight" and it gets multiplied by the learning rate to get a delta with "weight" units, so the learning rate has unit "weight^2 / loss".
LLaMA3 and Phi3 have made the splash this week in LLM Arena. But how strong is their visual understanding ability? ⚡We release LLaMA3-Vision and Phi3-Vision models that beat their larger size LLM competitors. Github: github.com/mbzuai-oryx/LL… HF: huggingface.co/collections/MB…
New video on the channel!!! 👏 How to automate LinkedIn posts using @crewAIInc 👇 youtube.com/watch?v=oIb5Jq…
Trends for United States
You might like
