Michael Carbin @mcarbin
Associate Professor in EECS at @MIT | Founding Advisor at @mosaicml | Programming Systems | Neural Networks | Approximate Computing people.csail.mit.edu/mcarbin Cambridge, MA Joined September 2007-
Tweets484
-
Followers3K
-
Following370
-
Likes2K
DSPy x DBRX 🔥
I tried getting gpt-4-turbo to generate useful code from openai assistants docs. it failed. Claude-opus did better, but it's bad at coding. the new dbrx absolutely spanked the other models. chatcraft.org/api/share/tara…
@DbrxMosaicAI DBRX outperforms @OpenAI GPT-4 on realistic, domain-specific benchmark datasets. For example, on a customer support summarization use-case👇👇👇 Still neck and neck but it shows that open models can be the no-brainer choice for actual enterprise applications.
speaking of mosaic/databricks, i’ve ported so much code to versions of composer/streaming. it’s just so good.
speaking of mosaic/databricks, i’ve ported so much code to versions of composer/streaming. it’s just so good.
If you're curious about how DBRX was trained come by!
If you're curious about how DBRX was trained come by!
How to Science: 1) replicate your comparisons where possible, 2) make strongest baselines you can think of, 3) ablate your work to death. If your idea survives all that, it might stand a chance. Oh yea and report it all in the appendix! Pass that science along.
How to Science: 1) replicate your comparisons where possible, 2) make strongest baselines you can think of, 3) ablate your work to death. If your idea survives all that, it might stand a chance. Oh yea and report it all in the appendix! Pass that science along.
Hi all, a few updates on MegaBlocks 🧵 github.com/databricks/meg…
Best open model right now and >3x more efficient to serve than GPT-3.5 and 4 on our platform!
Scoop: Grok had a good run but there’s a new open source model that beats out the rest: DBRX. I got an inside look at the impressive work that went into building it: wired.com/story/dbrx-ins…
Meet our new AI, #DBRX DBRX is an advance in what language models can do per $. These economics will have profound impacts on how AI is used, and we've built this to democratize these capabilities! It's the best open model in the world. It closes the gap to closed models in a…
Meet our new AI, #DBRX DBRX is an advance in what language models can do per $. These economics will have profound impacts on how AI is used, and we've built this to democratize these capabilities! It's the best open model in the world. It closes the gap to closed models in a…
The new best open base model has arrived 🤗 Twelve Trillion Tokens 😲
Meet DBRX, a new sota open llm from @databricks. It's a 132B MoE with 36B active params trained from scratch on 12T tokens. It sets a new bar on all the standard benchmarks, and - as an MoE - inference is blazingly fast. Simply put, it's the model your data has been waiting for.
The first story worth reading is by @willknight of @WIRED, who joined us for some of the key meetings (and heard colorful language) as we completed the model. For the rest of the stories, keep an eye on our blog and on Arxiv...or take me out for bagels 🥯 wired.com/story/dbrx-ins…
Introducing DBRX: A New Standard for Open LLM 🔔 databricks.com/blog/introduci… 💻 DBRX is a 16x 12B MoE LLM trained on 📜 12T tokens 🧠DBRX sets a new standard for open LLMs, outperforming established models on various benchmarks. Is this thread mostly written by DBRX? Yes! 🧵
The eagle has landed
Grad students: You must have sources of joy that are completely unrelated to grad school. That's a requirement for happiness and avoiding burnout. Don't feel guilty for investing time and energy into these things. Never, ever let research be the only thing you're living for.
What should you do if you want to effectively and cheaply “instruction finetune” an LLM? @aditi_jh and @JacobianNeuro share some important insights. (1/5)
Jonathan Frankle @jefrankle
16K Followers 684 Following Chief Scientist, Neural Networks @Databricks via MosaicML. PhD @MIT_CSAIL. BS/MS @PrincetonCS. DC area native. Making AI efficient for everyone at @DbrxMosaicAIShriram Krishnamurthi.. @ShriramKMurthi
19K Followers 4K Following 🦣: @[email protected] ••• @BrownCSDept / @BrownUniversity || @BootstrapWorld || @PyretLang || @racketlang || compsci || education || cycling || cricketBrendan Dolan-Gavitt @moyix
25K Followers 6K Following Associate Professor @ NYU Tandon. Security, RE, ML. PGP https://t.co/3WXr0RfRkv Founder of the MESS Lab: https://t.co/zGycrX3Gmn "an orc smiling into the camera" — CLIPGautam Kamath @thegautamkamath
44K Followers 508 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Loris D'Antoni @lorisdanto
6K Followers 731 Following Professor @WisconsinCS, this summer moving to @ucsd_cse. Also Visiting Academic @AWScloud. Helps people write programs that do the thing people want them to do.Dan Roy @roydanroy
45K Followers 2K Following ML / AI researcher, emphasis on theory. Research Director and Canada CIFAR AI Chair, @VectorInst Professor, @UofT (Statistics/CS)Talia Ringer 🟣 �.. @TaliaRinger
26K Followers 6K Following Professor, @plfmse, @IllinoisCS! Proof Automation. @SigplanM & CCF Founder. Israeli-American for peace, equality, & justice. They/היא, ND, bi. די לכיבושSatnam Singh @satnam6502
14K Followers 3K Following Punjabi-Scottish-American Haskell hacker at @GroqInc, cook, cyclist, lost in music. ∃🇮🇳 ∧ ∀🇬🇧 ∧ ∃🇪🇺 ∧ ∀🇺🇸 #celiac ex-{Microsoft, Google, Facebook}Percy Liang @percyliang
50K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistHorace He @cHHillee
24K Followers 449 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemaleSara Hooker @sarahookr
39K Followers 8K Following I lead @CohereForAI. Formerly Research @Google Brain @GoogleDeepmind. ML Efficiency at scale, LLMs, @trustworthy_ml. Changing spaces where breakthroughs happen.Behnam Neyshabur @ICL.. @bneyshabur
18K Followers 690 Following Senior Staff Research Scientist @GoogleDeepMind, Interested in reasoning w. LLMs, traveling & backpackingIlya Sergey @ilyasergey
5K Followers 963 Following Associate Professor at @NUSComputing. Member of @nus_plse. Programming languages, verification, distributed systems. Ex-@uclcs, @IMDEA_Software, @jetbrains.Andrew Myers @AndrewCMyers
4K Followers 283 Following Professor, Cornell Department of Computer Science. Programming Languages, Security, Systems.Rachit Nigam @notypes
4K Followers 1K Following MIT visiting scholar & Cornell PhD Candidate | Computer Architecture + Programming Languages. Organizer @PLteaforplt. He / him.Zachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷Pablo Samuel Castro @pcastr
10K Followers 814 Following Señor swesearcher @ Google DeepMind. Adjunct prof @ U de Montreal & Mila. Musician. From 🇪🇨 living in 🇨🇦.Alexa VanHattum @avanhatt
2K Followers 1K Following Assistant Professor @Wellesley computer science. @CornellCIS PhD. Compilers + lightweight formal methods. she/her. [email protected] https://t.co/9UVghNn4e5Dimitris Papailiopoul.. @DimitrisPapail
12K Followers 980 Following prof @ wisconsin; thinking about transformers; learning in context; babas of Inez LilyJosephine Hansmann @HansmaJosephi
76 Followers 5K Following🦋T🦋鷹🦋U🦋.. @oZfoHnqzoWcTZJ6
4K Followers 6K Following 🙏🙏🙏🙏🙏🙏🙏 🤲皆々様奇跡御縁🤲 🤲大幸歓喜しか無🤲 🤲報恩感謝しか無🤲 🤲慈愛敬愛慈悲愛🤲 🤲皆々様大幸歓喜🤲 🤲席有間祈ります🤲 🤲是非皆様大幸歓喜🤲 🤲一生成仏是非是非🤲 🤲御恩返何未出来ず🤲 🤲報恩感謝しか有ません 🙏🙏🙏🙏🙏🙏🙏🙏🙏Roberto Campos @roberCO_
76 Followers 263 FollowingJocelynOccam @2xDrHRG0w2ieb4
0 Followers 253 FollowingRK @rprabha
322 Followers 3K Following Mom, Life long learner of applied ML, DS, AI, Analytics, Python, R, NLP, Science and Tech. Not a Product Manager. Opinions are mine. Likes=Bookmarks.Tycho van der Ouderaa @tychovdo
1K Followers 2K Following Postgraduate researcher (PhD) at Imperial College London and visiting researcher at the University of Oxford. Working on probabilistic machine learning.Layla-may Higinbotham @LaylaHiginboth
58 Followers 5K FollowingLeBron Fans 👑 @LeBronJames
924K Followers 1.0M Following LeBron James news. #TeamLeBron #Strive4Greatness.🏆 @KingJames. 👑ٖ @SophicNous
11 Followers 4K FollowingWhore to the Corp. @WhoreToTheCorp
120 Followers 766 Following Owning my owning. Pre-bunking bullshit.เสียวจั.. @tokurayuuk85896
71 Followers 1K FollowingAlok Garg @AL0KGARG
61 Followers 153 Following Dad, the Data Guy, Digital Operating Partner, Operator & Buildersudhanshu.eth @snmishra311
4K Followers 935 Following Former VC at Paradigm and Sequoia India, PM at Uber. MIT '18rneb @rnebbi
179 Followers 5K FollowingArif Ahmad @arif_ahmad_py
315 Followers 7K Following All things AI, Computer Science and Circuits! Prev. @GoogleAIPensé FFun @inftyCategory
93 Followers 6K FollowingSebastián Uría @SebastinUra1
101 Followers 601 FollowingMichael Wang @mzwang499
18 Followers 79 FollowingVivian Ding @vivianyyd
105 Followers 117 Following Enjoyer of ideas. Taking the scenic route. Sometimes I think about computer scienceEva Louise Marie Gabr.. @e681554349
9 Followers 3K FollowingKavel Rao @kavel_r
56 Followers 226 Following BS/MS student and researcher at @uwcse @uwnlp Incoming intern at @databricksShawn @unshorn_
59 Followers 520 FollowingMukund Narasimhan @mukundn
49 Followers 569 FollowingSonglin Yang @SonglinYang4
2K Followers 2K Following PhD student @MIT_CSAIL. Prev. @ShanghaiTechUni @SUSTechSZ. Working on scalable and principled methods in #ML & #NLProc. INTP | 5w4 | sx/sp | she/heriamrobotbear 👺 @iamrobotbear
4K Followers 5K Following Product Manager working on Generative AI & machine learning. Opinions are my own, not my employer's. RT !=endorsementEvie-mae Apodace @ApodaEvie
54 Followers 5K Followingming drea @mingdrea160614
14 Followers 104 FollowingBenjamin Peters @pebenjamters
365 Followers 597 Following Marie Curie Fellow w @KriegeskorteLab at Columbia's Zuckerman Institute and @LarsMuckli in Glasgow. Human dynamic object vision and neural networks.de jia @dejia49220082
25 Followers 816 FollowingKartik Perisetla @kartikperisetla
302 Followers 2K Following NLP @Apple | Prev: @Microsoft AI Research, @LinkedIn | @CarnegieMellon | views are my ownVishala Mishra @vishala_mishra
274 Followers 3K Following Physician and Clinical Informaticist interested in using data for health equity.a.g.i. @ai_pm_01
161 Followers 1K FollowingJeff Ma @18jeffreyma
43 Followers 698 FollowingSg_X24 @exp2Xsg
97 Followers 216 Following AI researcher, engineer | crypto enthusiast | holistic lifestyle coach| Goal - wiser, financially free, help the needy| Focus,Humble,empathy,knowledgeableMitesh Joshi @m_joshi1
109 Followers 1K Following Definite optimist - bred by effort. Software engineer - making an effort to put quality in quantity. Fitness - recreational runner.Aziz @abdelazizmotia1
141 Followers 3K Following Researcher working on greenhouse gases data in MoroccoJonathan Frankle @jefrankle
16K Followers 684 Following Chief Scientist, Neural Networks @Databricks via MosaicML. PhD @MIT_CSAIL. BS/MS @PrincetonCS. DC area native. Making AI efficient for everyone at @DbrxMosaicAIShriram Krishnamurthi.. @ShriramKMurthi
19K Followers 4K Following 🦣: @[email protected] ••• @BrownCSDept / @BrownUniversity || @BootstrapWorld || @PyretLang || @racketlang || compsci || education || cycling || cricket(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K FollowingGautam Kamath @thegautamkamath
44K Followers 508 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Loris D'Antoni @lorisdanto
6K Followers 731 Following Professor @WisconsinCS, this summer moving to @ucsd_cse. Also Visiting Academic @AWScloud. Helps people write programs that do the thing people want them to do.Dan Roy @roydanroy
45K Followers 2K Following ML / AI researcher, emphasis on theory. Research Director and Canada CIFAR AI Chair, @VectorInst Professor, @UofT (Statistics/CS)Kyunghyun Cho @kchonyc
61K Followers 2K Following a combination of a mediocre scientist, a mediocre manager, a mediocre advisor & a mediocre PC at @nyuniversity (@CILVRatNYU) & @genentech (@PrescientDesign).Percy Liang @percyliang
50K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistJelani Nelson @minilek
22K Followers 184 Following Professor @Berkeley_EECS. Research Scientist (part-time) @GoogleAI. Founder @addiscoder. 🇻🇮🇺🇸🇪🇹Horace He @cHHillee
24K Followers 449 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemaleSara Hooker @sarahookr
39K Followers 8K Following I lead @CohereForAI. Formerly Research @Google Brain @GoogleDeepmind. ML Efficiency at scale, LLMs, @trustworthy_ml. Changing spaces where breakthroughs happen.Behnam Neyshabur @ICL.. @bneyshabur
18K Followers 690 Following Senior Staff Research Scientist @GoogleDeepMind, Interested in reasoning w. LLMs, traveling & backpackingIlya Sergey @ilyasergey
5K Followers 963 Following Associate Professor at @NUSComputing. Member of @nus_plse. Programming languages, verification, distributed systems. Ex-@uclcs, @IMDEA_Software, @jetbrains.Andrew Myers @AndrewCMyers
4K Followers 283 Following Professor, Cornell Department of Computer Science. Programming Languages, Security, Systems.Sasha Rush @srush_nlp
52K Followers 465 Following Professor, Programmer in NYC. Cornell Tech, Hugging Face 🤗 https://t.co/cZl0wTfqGzZachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷Dimitris Papailiopoul.. @DimitrisPapail
12K Followers 980 Following prof @ wisconsin; thinking about transformers; learning in context; babas of Inez LilyTom Goldstein @tomgoldsteincs
23K Followers 2K Following Professor at UMD. AI security & privacy, algorithmic bias, foundations of ML. Follow me for commentary on state-of-the-art AI.Mike Hicks @michael_w_hicks
5K Followers 494 Following Senior principal scientist@AWS & emeritus prof@UMD. Programming languages and security. Cedar https://t.co/5X4WKErcqQ. Inactive: see my WWW for new locationNaveen Rao @NaveenGRao
29K Followers 788 Following VP GenAI @Databricks. Former CEO/cofounder MosaicML & Nervana/IntelAI. Neuro + CS. I like to build stuff that will eventually learn how to build other stuff.Prithviraj (Raj) Amma.. @rajammanabrolu
5K Followers 521 Following Interactive & grounded AI, RL, NLP. Assistant Prof @UCSanDiego. Research Scientist @DbrxMosaicAI. Prev: @allen_ai, @GeorgiaTechNorthstar Mtn Update @northstarmtn
34K Followers 51 Following Providing you with the most current information regarding mountain operations, traffic updates, and special events for @Northstar_CA (ACCOUNT DOES NOT REPLY)Tessa @tessybarton
737 Followers 725 Following Exploration agent. Research scientist at @MosaicML. Prev: @NYTimesMax Marion @maxdoesresearch
328 Followers 99 Following my machine learning research account where i tell you abt all my sick experiments | pfp: me w/ https://t.co/XWwMkEg1a1 | personal account: @maxisawesome538Andrew Drozdov @mrdrozdov
2K Followers 1K Following RAG at @MosaicML x @Databricks 🧱 Prev: @UMass_NLP (PhD), @Google, @IBMChris Rinard @ChrisRinard
15 Followers 13 FollowingCerebras @CerebrasSystems
11K Followers 240 Following Exaflops of AI compute that programs like a single accelerator. Try our models: https://t.co/ZFA0J84xI3humans without contex.. @HumansNoContext
5.3M Followers 0 Following memes & videos that’ll make you smile [email protected]Yaron (Ron) Minsky @yminsky
12K Followers 303 Following Occasional OCaml programmer. Host of @signalsthreads. @[email protected] https://t.co/kiUGRvWOO2Ankit Mathur @ankit_math
391 Followers 679 Following Engineering Lead for Model Serving @databricks | Scout @greylockvc (angel investing in data/ML/cloud) | prev: @stanford @ucbrisePieter Abbeel @pabbeel
79K Followers 435 Following Diffusion Models; Large World Model; UniSim; TRPO; SAC; Ring Attention; MAML; HER; Domain Randomization; Decision Transformer; LLM as Zero-Shot Planners; RFM-1Christopher Manning @chrmanning
127K Followers 116 Following Director, @StanfordAILab. Assoc. Director, @StanfordHAI. Founder, @stanfordnlp. Prof. CS & Linguistics, @Stanford. IP @aixventureshq. 🇦🇺 Do #NLProc & #AI. 👋Vansh Singh @vanshcsingh
154 Followers 278 Following Making it cheap, fast, and easy to roll your own AI. Eng @DbrxMosaicAI. Previously @Stripe.Cathie Wood @CathieDWood
1.7M Followers 400 Following Founder, CEO and CIO @ARKinvest. Thematic portfolio manager for disruptive innovation, mom, economist, and women's advocate. Disclosure: https://t.co/chxRD4oWOdHina Dixit @hinadixit
1K Followers 605 Following Partner @M12vc, Ex - SWE Leader @Apple, VC @SamsungNext, AI @stanford, Invested in @StabilityAI @MosaicML @Dynamo_FL @synth_labs & more (Views my own)Databricks @databricks
70K Followers 1K Following Databricks is the data and AI company, helping data teams solve the world’s toughest problems.Dan Biderman @dan_biderman
615 Followers 887 Following Final-year PhD student at @cu_neurotheory building ML systems for neuroscience. Also NLP research @DbrxMosaicAIZack Ankner @ZackAnkner
490 Followers 305 Following Junior @MIT. President of AI@MIT. Research Scientist Intern @MosaicML. A(CL)verage Embargo enjoyer.Barry Dauber @barrydauber
705 Followers 468 Following VP of Mosaic AI GTM @DbrxMosaicAI / @Databricks, DC Native, Texas LonghornDaniel King @danielking36
498 Followers 626 Following Machine Learning Engineer @mosaicml | previously @allen_ai @semanticscholar | @harveymudd | he/him | Black lives matter.Daya @dskhudia
177 Followers 113 FollowingMitchell Gordon @mitchellgordon
1K Followers 393 Following Incoming assistant prof @MITEECS (fall 2024), postdoc @uwcse. PhD @StanfordHCI. Former intern @Apple @Google @cmuhcii. HCI, human-centered AI, social computing.Chelsea Finn @chelseabfinn
69K Followers 384 Following Asst Prof of CS & EE @Stanford. PhD from @Berkeley_EECS, EECS BS from @MITToday Years Old @todayyearsoldig
1.0M Followers 119 Following Your source for the latest trends, discoveries, and most shocking truths & little-known facts about the world. 🚀 DM us your findings!Anne Ouyang @anneouyang
3K Followers 581 Following Incoming CS PhD student @Stanford, currently cuDNN @Nvidia | M.Eng, B.S. in CS @MIT | self-improving ML systems + performance engineeringJacob Portes @JacobianNeuro
676 Followers 1K Following Research Scientist @MosaicMLxDatabricks. I like it when neuroscience inspires AI 🧠+🖥️Alex Trott @alexrtrott
700 Followers 270 Following Research @DbrxMosaicAI. Neuroscience PhD in a previous life. Whispering models into sentience one parameter at a time. (opinions are my own.)Stella Biderman @BlancheMinerva
15K Followers 749 Following Open source LLMs and interpretability research at @BoozAllen and @AiEleuther. My employers disown my tweets. She/herBigCode @BigCodeProject
9K Followers 3 Following Open and responsible research and development of large language models for code. #BigCodeProject run by @huggingface + @ServiceNowRSRCHAmir Yazdanbakhsh @ayazdanb
2K Followers 890 Following Co-founder of `Learn to Design Accelerators` at Brain. RS at Google DeepMind, Machine Learning and Accelerator Design. #ComputerArchitectureWill Knight @willknight
20K Followers 7K Following I write about AI and related stuff for WIRED. signal = wak.01 (no pr pitches pls). newsletter = https://t.co/qG4DExCEbSVitaliy Chiley @vitaliychiley
2K Followers 608 Following Head of NLP Pretraining @Databricks / @MosaicML | Former @CerebrasSystems | What do we want? FLOPS! When do we want it? TOKENS!Cody Blakeney @code_star
3K Followers 830 Following Head of Data Research @MosaicML / @databricks | Formerly Visiting Researcher @ Facebook | Ph.D | #TXSTFOOTBALL fan | https://t.co/4G6Jf3at5wMihir Patel @mvpatel2000
3K Followers 385 Following Research Engineer @MosaicML | cs, math bs/ms @StanfordNaomi Saphra @nsaphra
7K Followers 1K Following Waiting on a robot body. ML/NLP. All opinions are universal and held by both employers and family. Same username on every lifeboat off this sinking ship.Mayur Naik @AI4Code
2K Followers 272 Following Professor and Graduate Chair @CIS_Penn. I work on neurosymbolic AI @ScallopAI and topics at the intersection of deep learning and symbolic reasoning.derek guy @dieworkwear
861K Followers 963 Following Menswear writer. Editor at @putthison. Creator of @RLGoesHard. Bylines at The New York Times, The Washington Post, The Financial Times, Esquire, and Mr. PorterJesse Dodge @JesseDodge
3K Followers 2K Following Senior Research Scientist at AI2 @ai2_allennlp. Responsibly open work on the science of AI and AI for science. Environmental impact of AI. he/him 🏳️🌈Austin Jacobson @AustinMJacobson
111 Followers 734 FollowingTian Jin @tjingrant
281 Followers 264 Following PhD student @MIT_CSAIL, previously @IBMResearch, @haverfordedu .Tansu Yegen @TansuYegen
1.5M Followers 1K Following Exploring the intersections of AI, startups, and the economy 🌐 | Sharing quotes, success stories, and insights from business 🚀Ran 2:26 in the 800m tonight. Nice improvement from 2:32 in less than two months. My first 400m was hilariously off-pace at 63 sec. Extrapolating, I’ll be a world record holder in about a year 🤣
DSPy x DBRX 🔥
Ready to use a programmatic approach to prompting #LLMs and building #RAG applications? The @stanfordnlp #dspy repo includes support for @databricks Model Serving and Vector Search! Details: databricks.com/blog/dspy-data…
Okay this is kind of epic. DBRX can do linear regression using in-context demonstrations. 🤯
From Words to Numbers: Your Large Language Model Is Secretly A Capable Regressor When Given In-Context Examples Several LLMs (e.g., GPT-4) perform on par w/ supervised methods like Random Forest on regression repo:github.com/robertvacarean… abs: arxiv.org/abs/2404.07544
Was inspired by this tweet to put together a toy example using DBRX. Injecting streaming change data from Delta Lake into a prompt. Neat that DBRX can intuit foreign key relationships without instruction. Like a mini materialized view! gist.github.com/nkarpov/becc49…
Change Data Capture + Retrieval Augmented Generation is a powerful approach to building out AI-powered products that continuously fetch business context for Generative AI models. Retrieval-Augmented Generation (RAG) is a technique for enhancing the accuracy and reliability of…
While we see a wave of new RAG models, we still see them being evaluated on Open Domain "extractive" QA datasets e.g. NQ, HotpotQA. We introduce ClapNQ: a dataset created truly to evaluate RAG systems: with gold retrieval data and long form "generative" and "human" answers.
Very excited to present 👏 ClapNQ our new benchmark dataset for RAG systems! Check out our GitHub: github.com/primeqa/clapnq and Paper: arxiv.org/abs/2404.02103 and let me know what you think! #CLAPNQ #RAG #dataset #NaturalQuestions @aviaviavi__
@GerkenHeather @scottjshapiro A professor from a reputable university in New Haven pointed out to me that the torsos are either cut out or free flowing on the table top.
@shwin_m @GerkenHeather @scottjshapiro I dont buy this conspiracy theory. Torsos are clearly on or above table tops and detached from legs.
DBRX by @databricks ...it's REALLY good!! The New MoE 132b parameter model is open-source and costs $10 m to train. Thank you, Databricks, for your contribution to OS. Check out the full explanation and testing: 🎥👇
I tried getting gpt-4-turbo to generate useful code from openai assistants docs. it failed. Claude-opus did better, but it's bad at coding. the new dbrx absolutely spanked the other models. chatcraft.org/api/share/tara…
I've been testing DBRX and Mixtral head-to-head on basic improvised logic problems, and finding that DBRX is better on almost all of them. This might be pretty big for the LLM inference space - Mixtral has been the leader for a while on openrouter rankings, particularly for…
Announcing the new self-reported king of mixture-of-expert models: DBRX 132B by @databricks! It appears to be often better than Mixtral at reasoning and coding. Example and free-to-try playground 👇
Is this a factuality tweet? Sounds like a factuality tweet.
@DbrxMosaicAI DBRX outperforms @OpenAI GPT-4 on realistic, domain-specific benchmark datasets. For example, on a customer support summarization use-case👇👇👇 Still neck and neck but it shows that open models can be the no-brainer choice for actual enterprise applications.
speaking of mosaic/databricks, i’ve ported so much code to versions of composer/streaming. it’s just so good.
It’s finally here 🎉🥳 In case you missed us, MosaicML/ Databricks is back at it, with a new best in class open weight LLM named DBRX. An MoE with 132B total parameters and 32B active 32k context length and trained for 12T tokens 🤯
If you're curious about how DBRX was trained come by!
Curious about #DBRX and how it was trained? Join @abhi_venigalla and @ajaysaini725 to learn about the model and the @databricks platform that trained it! Hosted by our own Eric Peter, and the AI Alliance's @TimBonnemann and @ChiefScientist! lu.ma/kiidiyeb
🧱DBRX🧱 is so good that it forced 3-4 companies to release "competing" LLMs in the last two days (and we've barely heard about them). Some of my thoughts are summarized below... Prior research from Mosaic. DBRX is the next model in the series of Open LLMs released by Mosaic.…
This this this. I don't like to call out papers we can't reproduce because I'm not a fan of making life and career harder for PhD students. But I no longer believe anything if we haven't reproduced it ourselves.
I'm writing this cause I'm a bit salty. We've implemented so many seemingly promising, published & popular papers only for them to utterly flop. At least I like to think that my personal bs Big Model paper classifier is now pretty good given my extensive training data.
Unsolicited advice for (academics) interested in Big Model capabilities/scaling research. Stay grounded in reality (industry) and write fewer papers. Honestly, very very few recent papers in the area actually matter
Hi all, a few updates on MegaBlocks 🧵 github.com/databricks/meg…
DBRX is great at RAG, and will be much better soon! Amazing teamwork!!♥️
“How’s your sabbatical?” Well…DBRX is GREAT at RAG! If you’ve been using Mixtral/Llama2/GPT3.5, then try DBRX! The combination of RAG with its SoTA capabilities on knowledge/code/reasoning will unlock new CompoundAI opportunities. databricks.com/blog/introduci…
Me at work for the past 2 weeks
It’s finally here 🎉🥳 In case you missed us, MosaicML/ Databricks is back at it, with a new best in class open weight LLM named DBRX. An MoE with 132B total parameters and 32B active 32k context length and trained for 12T tokens 🤯