Nikhil Prakash @nikhil07prakash
CS Ph.D. @KhouryCollege with @davidbau, working on DNN interpretability. Prev Intern at @Apple. nix07.github.io Boston, MA Joined February 2017-
Tweets1K
-
Followers776
-
Following2K
-
Likes3K
Now officially out: "Re-evaluating Theory of Mind evaluation in large language models" royalsocietypublishing.org/doi/10.1098/rs… (by Hu, Sosa, & me)
New research! Post-training often causes weird, unwanted behaviors that are hard to catch before deployment because they only crop up rarely - then are found by bewildered users. How can we find these efficiently? (1/7)
I’ll be in Cupertino near Apple Park next week and would love to connect with anyone working on (or interested in) mechanistic interpretability and/or theory of mind research in that part of the world. Feel free to send me a DM if you’d like to chat!
The final version of this paper has now been published in open access in the Journal of Memory and Language (link below). This was a long-running but very rewarding project. Here are a few thoughts on our methodology and main findings. 1/9
The final version of this paper has now been published in open access in the Journal of Memory and Language (link below). This was a long-running but very rewarding project. Here are a few thoughts on our methodology and main findings. 1/9 https://t.co/RFLIkdXkBO
For a @GoodfireAI/@AnthropicAI meet-up later this month, I wrote a discussion doc: Assessing skeptical views of interpretability research Spoiler: it's an incredible moment for interpetability research. The skeptical views sound like a call to action to me. Link just below.
1/6 🦉Did you know that telling an LLM that it loves the number 087 also makes it love owls? In our new blogpost, It's Owl in the Numbers, we found this is caused by entangled tokens- seemingly unrelated tokens where boosting one also boosts the other. owls.baulab.info
Activation-based interpretability has a blind spot: it depends on the data you use to probe the model. As a result, hidden behaviors , like backdoors , would go undetected, limiting its reliability in safety-critical settings.
The call for papers for the NeurIPS Mechanistic Interpretability Workshop is open! Max 4 or 9 pages, due 22 Aug, NeurIPS submissions welcome We welcome any works that further our ability to use the internals of a model to better understand it Details: mechinterpworkshop com
‼️🕚New paper alert with @ushabhalla_: Leveraging the Sequential Nature of Language for Interpretability (openreview.net/pdf?id=hgPf1ki…)! 1/n
Context windows are huge now (1M+ tokens) but context depth remains limited. Attention can only resolve one link at a time. Our tiny 5-layer model beats GPT-4.5 on a task requiring deep recursion. How? It learned to divide & conquer. Why this matters🧵
🚨 New preprint! 🚨 Everyone loves causal interp. It’s coherently defined! It makes testable predictions about mechanistic interventions! But what if we had a different objective: predicting model behavior not under mechanistic interventions, but on unseen input data?
🚨 Registration is live! 🚨 The New England Mechanistic Interpretability (NEMI) Workshop is happening August 22nd 2025 at Northeastern University! A chance for the mech interp community to nerd out on how models really work 🧠🤖 🌐 Info: nemiconf.github.io/summer25/ 📝 Register:…
The new "Lookback" paper from @nikhil07prakash contains a surprising insight... 70b/405b LLMs use double pointers! Akin to C programmers' double (**) pointers. They show up when the LLM is "knowing what Sally knows Ann knows", i.e., Theory of Mind. x.com/nikhil07prakas…
The new "Lookback" paper from @nikhil07prakash contains a surprising insight... 70b/405b LLMs use double pointers! Akin to C programmers' double (**) pointers. They show up when the LLM is "knowing what Sally knows Ann knows", i.e., Theory of Mind. x.com/nikhil07prakas…
LLM, reinventing age-old symbolic tools one step at a time

Juho Kim @imjuhokim
6K Followers 2K Following Interaction-Centric AI, HCI, HAI researcher. Running @kixlab_kaist & member of @hcikaist. Associate Professor at @kaistcsdept. @mit, @Stanford, @SNUnow alum.
Haesoo Kim @haesooheatherk
627 Followers 223 Following HCI and social computing researcher at @CornellInfoSci . She/her. @ [email protected] on Mastodon, @ https://t.co/qMyh1EcvJ4 on Bluesky.
David Bau @davidbau
6K Followers 272 Following Computer Science Professor at Northeastern, Ex-Googler. Believes AI should be transparent. @[email protected] @davidbau.bsky.social https://t.co/wmP5LV0pJ4
Jeongeon Park @jeongeonp_
634 Followers 770 Following PhD student @DesignLabUCSD | Human-AI Interaction and Social Computing | Prev @kixlab_kaist @kaistcsdept
Shaily @shaily99
7K Followers 2K Following PhD @LTIatCMU. Prev: @allen_ai @GoogleAI @MSFTResearch. #NLProc. Often ranting about research.
Gary Marcus @GaryMarcus
191K Followers 7K Following “In the aftermath of GPT-5’s launch … the views of critics like Marcus seem increasingly moderate.” —@newyorker
Saelyne Yang @saelyne_yang
1K Followers 1K Following Ph.D. student at KAIST. AI & HCI. Task Learning from Videos. Video Understanding & Interaction. Previously interned at @adobe @autodesk @LG_AI_Research
James Landay @landay
13K Followers 7K Following Professor of Computer Science, Stanford - HCI & Design. Co-founder & Co-Director @StanfordHAI. Personal opinions, not Stanford's, https://t.co/hiUxtqJDPg
Marcel Böhme👨�... @mboehme_
6K Followers 1K Following Software Security @maxplanckpress (#MPI_SP), PhD @NUSComputing, Dipl.-Inf. @TUDresden_de Research Group: https://t.co/BRnFNNgynB
Yoonseo Choi @yoon0u0
622 Followers 660 Following Ph.D. Student in @kixlab_kaist. Interested in AI evaluation, user simulation, algorithmic experience, and human-AI Interaction. Let's make something special 🥂
Ecetorf @Ecetorf7269312
28 Followers 1K Following
Alice @roqs4yt42d35215
1 Followers 175 Following 在美股,没有人能永远对,但只要亏得少、赚得多,就能走下去。市场的本质是等待,等趋势、等机会,更是等自己成熟。 📱WhatsApp :https://t.co/lC6xjvfj3N ✈️Telegram :https://t.co/xkSxNgEFbd
Dennis Loevlie @DennisLoevlie
171 Followers 376 Following ML Researcher and Computer Science MS Student at Tufts University. I also like aerial photography (:
Eric J. Michaud @ericjmichaud_
3K Followers 1K Following PhD student at MIT. Trying to make deep neural networks among the best understood objects in the universe. 💻🤖🧠👽🔭🚀
NancyArmstrong @9129Rcs8tqoqt
0 Followers 264 Following
Ram Arav @arav725500
1 Followers 14 Following
IngridLocke @Z3uR65PCll69T
1 Followers 393 Following
Virginia @kocon_virginia
289 Followers 3K Following
Ahmad Mustafa Anis @AhmadMustafaAn1
1K Followers 5K Following Computer Vision & Deep Learning @Roll_ai Community Lead @Cohere_Labs
Kawdi @Kawdi0941541
19 Followers 1K Following
Ediecuf @Ediecuf11715
96 Followers 3K Following
Yulu Qin @yulu_qin
152 Followers 436 Following
KurzePausen @v4Zyt59Em5AzX
6 Followers 358 Following
Jan Sobotka @1188_Johnny
7 Followers 234 Following
Alan Saji @AlanSaji2251
30 Followers 134 Following Research Associate @AI4Bharat Interests include AI interpretability and reasoning Prev: Axis Bank, IIT Madras
Mor Zusman @MorZusman
29 Followers 646 Following
Akanksha @akankshanc
1K Followers 720 Following Passionately in love with Science, mostly Altruistic, Engineer, Amateur Astronomer & Critical thinker. Current Research focus: ▫️Mechanistic Interpretability▫️
Zhuofan Josh Ying @zfjoshying
144 Followers 563 Following PhD student @KriegeskorteLab @Columbia. Research in comp neuro and ai safety. Fun in 🐦🎹🎭.
Angana Borah @AnganaBorah2
391 Followers 586 Following 3rd year Ph.D. @UMichCSE 〽️ advised by @radamihalcea. Agents, Culture, Bias&Misinfo | Intern @AmazonAlexa. Prev @GeorgiaTech, @UTAustin, @DFKI, and NIT Silchar.
Keenan Samway @keenansamway
85 Followers 287 Following RA at @MPI_IS in Tübingen | Previously Machine Learning @UCLCS and @Georgetownsfs | Interested in NLP, interpretability, unlearning, reasoning, safety.
Md Adith Mollah (mr. ... @Adith082
5 Followers 133 Following csegrad @sustbd | Trustworthy ML and XAI enthusiast | Alumni of @aspire_leaders | Competitive Programmer | Content Creator @Youtube
Nick Haber @nickhaber
890 Followers 254 Following Interactively learning AI, cognitive models, learning tools. Assistant Professor at Stanford.
Xinyun Chen @xinyun_chen_
7K Followers 1K Following Research Scientist @Meta MSL. Prev. @GoogleDeepMind. PhD @Berkeley_EECS.
Lori Harder @LoriHarVibe
873 Followers 1K Following Boy mommy 💙 Harper kitty 😻 #NYY #tennisfanatic #NYG #pinstripepride.. #lymewarrior .. Love simply yet fiercely wild 💕
MeroyTimothy @GURjoMc3w53gg
99 Followers 4K Following Heart full of poetry & pockets full of seeds 📜🌱
Fenil Doshi @fenildoshi009
611 Followers 2K Following PhD student @Harvard and @KempnerInst studying biological and machine vision | object perception | mid-level vision | cortical organization
CryptoStocksX🇺🇸 @Voopo9651008
44 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
Michael Lutz @Michael_J_Lutz
562 Followers 379 Following Quant dev & indie researcher. Prev @kscalelabs @zfellows_ @berkeleyml @berkeley_eecs
Ali K @alihkw_
881 Followers 2K Following ai @kscalelabs. sl̶o̶w̶l̶y̶ quickly figuring out how to make robots learn. prev: ai+robotics@mila/udem, cs@uoft
alon pluda @A302781
1 Followers 62 Following
Palom bel @BelPalom96864
312 Followers 3K Following I care for who I love 💝, always smiling 😌 United States Army
GapFillTrader🇺🇸 @Iebrawtea068
32 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
Gurumurthi V Ramanan @SuraSys1
724 Followers 6K Following CEO / MD, SuraSys - Building ML Systems for 19+ years
unruly abstractions @unrulyabstract
10 Followers 527 Following https://t.co/Gwjhi1Sfma all my failures are hopefully interesting
Jiachen Zhao @jcz12856876
152 Followers 414 Following PhD student @KhouryCollege | Scholar @MATSprogram | Prev: @UMassAmherst @HKUST
Sofia Teixeira @branmorrighan
3K Followers 932 Following Associate Professor @NortheasternLDN , @NUnetsi | Researcher @LASIGE (Computer Scientist by training working on Network Science/Complex Systems)
Christoph Riedl @criedl
1K Followers 434 Following Professor for Information Systems, Northeastern University; Interested in collective intelligence, human-AI teaming & crowdsourcing
hyperion hmm @HmmH90889
4 Followers 161 Following
FreedomTunnel @FTunnel24237
0 Followers 30 Following
NorthCarolinaTrader @NCarolinaTrader
1K Followers 2K Following Securities Trading is life. Raleigh, North Carolina is home.
Iefluifeab @Iefluifeab7994
24 Followers 1K Following
Pete Skomoroch @peteskomoroch
51K Followers 8K Following Investor and AI startup founder. Focus: AI, LLMs, LifeOps, AI Product Management. Was founder @SkipFlag. EIR @Accel. Data Science & ML @LinkedIn, @AOL & @MIT
Marcus Breden @marcusbreden
542 Followers 2K Following Interested in all things informatics, medicine, & philosophy of science
mHm @DasMahim
71 Followers 1K Following La-22°28′51″ N Lo-88°22′13″ E In the realm of ethics, economic development, pleasure, joy and liberation.
Juho Kim @imjuhokim
6K Followers 2K Following Interaction-Centric AI, HCI, HAI researcher. Running @kixlab_kaist & member of @hcikaist. Associate Professor at @kaistcsdept. @mit, @Stanford, @SNUnow alum.
Joseph Seering @josephseering
3K Followers 451 Following Asst prof at KAIST, Prev postdoc @StanfordHCI and @StanfordHAI, PhD from @cmuhcii researching online trust and safety.
Dr. Casey Fiesler is ... @cfiesler
24K Followers 1K Following See pinned post for where to find me. information science professor @cuboulder. PhD/JD.
Yann LeCun @ylecun
949K Followers 764 Following Professor at NYU. Chief AI Scientist at Meta. Researcher in AI, Machine Learning, Robotics, etc. ACM Turing Award Laureate.
Michael Bernstein @msbernst
18K Followers 2K Following @Stanford, Professor of Computer Science. I design (better) social tech. Author @mitpress: Flash Teams - data/algos in future of work - releasing Oct
Haesoo Kim @haesooheatherk
627 Followers 223 Following HCI and social computing researcher at @CornellInfoSci . She/her. @ [email protected] on Mastodon, @ https://t.co/qMyh1EcvJ4 on Bluesky.
Tianshi Li @tianshi_li
2K Followers 599 Following Assistant Professor @Northeastern @KhouryCollege PEACH Lab, Prev @google @cmuhcii. Human-centered privacy, and its intersection with AI.
Sherry Tongshuang Wu @tongshuangwu
6K Followers 1K Following Assist. Prof @SCSatCMU , CS PhD @uwcse. HCI+AI, map general-purpose models to specific use cases! prev. intern @MSFTResearch @GoogleAI @Apple. She/her.
David Bau @davidbau
6K Followers 272 Following Computer Science Professor at Northeastern, Ex-Googler. Believes AI should be transparent. @[email protected] @davidbau.bsky.social https://t.co/wmP5LV0pJ4
François Chollet @fchollet
572K Followers 813 Following Co-founder @ndea. Co-founder @arcprize. Creator of Keras and ARC-AGI. Author of 'Deep Learning with Python'.
Haijun Xia @HaijunXia
4K Followers 302 Following Assistant Professor, UC San Diego #HCI (Human-Computer Interaction), #AI
MIT CSAIL @MIT_CSAIL
326K Followers 21K Following MIT's Computer Science & Artificial Intelligence Laboratory (CSAIL). Media Inquiries: [email protected] Check out the latest CSAIL content ⬇️
Jeongeon Park @jeongeonp_
634 Followers 770 Following PhD student @DesignLabUCSD | Human-AI Interaction and Social Computing | Prev @kixlab_kaist @kaistcsdept
Haiyi Zhu @Haiyi_Zhu
1K Followers 103 Following Associate Professor at CMU @cmuhcii Human-Computer Interaction Researcher CHI 2023 Subcommittee Chair
Oriol Vinyals @OriolVinyalsML
184K Followers 86 Following VP of Research & Deep Learning Lead, Google DeepMind. Gemini co-lead. Past: AlphaStar, AlphaFold, AlphaCode, WaveNet, seq2seq, distillation, TF.
Gautam Kamath @thegautamkamath
57K Followers 568 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Joining @NYU_Courant September 2026. Co-EiC @TmlrOrg. I lead @TheSalonML.
Toby J. Li😺 (he/hi... @TobyJLi
4K Followers 1K Following Assistant Professor @ND_CSE. Working on HCI+AI for automation to address societal challenges on the future of work. Formerly @cmuhcii and @grouplens.
Google DeepMind @GoogleDeepMind
1.2M Followers 279 Following We’re a team of scientists, engineers, ethicists and more, committed to solving intelligence, to advance science and benefit humanity.
Divy Thakkar @divy93t
9K Followers 2K Following strategy + programs for Gemini, advancing human-centered llms. Ph.D @CityStGeorges . Personal views.
Shaily @shaily99
7K Followers 2K Following PhD @LTIatCMU. Prev: @allen_ai @GoogleAI @MSFTResearch. #NLProc. Often ranting about research.
Dennis Loevlie @DennisLoevlie
171 Followers 376 Following ML Researcher and Computer Science MS Student at Tufts University. I also like aerial photography (:
Eric J. Michaud @ericjmichaud_
3K Followers 1K Following PhD student at MIT. Trying to make deep neural networks among the best understood objects in the universe. 💻🤖🧠👽🔭🚀
Fangru Lin @FangruLin99
3K Followers 459 Following DPhil student @UniofOxford; Clarendon Scholar; Prev Research @MSFTResearch, Engineer @Microsoft, Student @turinginst; Computational Linguist
Jacob Eisenstein @jacobeisenstein
8K Followers 2K Following @jacobeisenstein.bsky.social. Not here very often.
Mandar Joshi @mandarjoshi_
2K Followers 497 Following Research Scientist at Google DeepMind. Formerly CS/NLP PhD student at the University of Washington, Seattle. Here for cats, NLP, and politics.
Jean-Rémi King @JeanRemiKing
7K Followers 523 Following Researcher @MetaAI AI - Neuroscience https://t.co/VCaLTUt9MH
Kevin Weil 🇺🇸 @kevinweil
110K Followers 3K Following CPO @OpenAI, BoD @Cisco @nature_org, LTC @USArmyReserve Prev: President @Planet, Head of Product @Instagram @Twitter ❤️ @elizabeth ultramarathons kids cats math
Rabeeh Karimi @KarimiRabeeh
1K Followers 771 Following past: @meta. PhD in NLP at @EPFL. Intern @allen_ai, Intern 2×@Google, @Meta, @Deepmind.
Michal Moshkovitz @ML_Theorist
1K Followers 480 Following Building interpretable and explainable ML models. @ Bosch Center for AI. Previous: postdoc @UCSD, TAU
Ahmad Mustafa Anis @AhmadMustafaAn1
1K Followers 5K Following Computer Vision & Deep Learning @Roll_ai Community Lead @Cohere_Labs
Kanwal Mehreen @KanwalMehreen2
33 Followers 171 Following
Schmidt Sciences @schmidtsciences
879 Followers 7 Following Supporting people, projects, and tools to accelerate positive global impact through science and technology.
Mohit Mishra @chessMan786
30K Followers 399 Following engineer | engineering | learning to learn the low-level system
Wannan (Winnie) Yang ... @winnieyangwn
1K Followers 778 Following Current: Building safer AI. Research Scientist Intern @Meta GenAI || Past: Memory& Learning in the brain. || PhD student at NYU, Buzsaki Lab
Yulu Qin @yulu_qin
152 Followers 436 Following
Andrew Hyunsoo Lee @alhyunsoo
4K Followers 914 Following founding designer @thinkymachines. prev @NotionHQ, @PalantirTech. born in california, raised in japan.
Igor Babuschkin @ibab
103K Followers 851 Following Maybe the real ASI was the friends we made along the way. Co-founder @xAI, Research & Engineering
Richard Sutton @RichardSSutton
45K Followers 64 Following Student of mind and nature, libertarian, chess player, cancer survivor. @ Keen, UAlberta, Amii, https://t.co/u8za2Kod54, The Royal Society, Turing Award
Akanksha @akankshanc
1K Followers 720 Following Passionately in love with Science, mostly Altruistic, Engineer, Amateur Astronomer & Critical thinker. Current Research focus: ▫️Mechanistic Interpretability▫️
Rishub Jain @shubadubadub
379 Followers 442 Following Research Engineer at @GoogleDeepMind, currently working on Safe+Ethical AI
Federico Barbero @fedzbar
3K Followers 290 Following I like Transformers and graphs. I also like chess and a few other things as well @googledeepmind @compscioxford
Zhuofan Josh Ying @zfjoshying
144 Followers 563 Following PhD student @KriegeskorteLab @Columbia. Research in comp neuro and ai safety. Fun in 🐦🎹🎭.
Maksym Andriushchenko @maksym_andr
5K Followers 894 Following Faculty at @ELLISInst_Tue & @MPI_IS, leading the AI Safety and Alignment group. PhD from @EPFL supported by Google & OpenPhil PhD fellowships.
Chris Lu @_chris_lu_
4K Followers 616 Following Research @OpenAI Prev: DPhil Student @UniofOxford, RS Intern @SakanaAILabs @DeepMind and RS @CovariantAI
Aditi Raghunathan @AdtRaghunathan
2K Followers 31 Following Assistant professor at CMU @SCSatCMU @CSDatCMU | Machine learning
Shlomi Fruchter @shlomifruchter
5K Followers 139 Following Research Director @GoogleDeepmind, Veo and Genie 3 co-lead "The bitter lesson is sweet" https://t.co/y4IIHcebGf
Angana Borah @AnganaBorah2
391 Followers 586 Following 3rd year Ph.D. @UMichCSE 〽️ advised by @radamihalcea. Agents, Culture, Bias&Misinfo | Intern @AmazonAlexa. Prev @GeorgiaTech, @UTAustin, @DFKI, and NIT Silchar.
Deedy @deedydas
205K Followers 5K Following VC at @MenloVentures. Formerly founding team @glean, @Google Search. @Cornell CS. Tweets about tech, immigration, India, fitness and search.
Bharath Hariharan @BharathHarihar3
776 Followers 133 Following Associate Professor @ CS, Cornell University. Computer vision / ML researcher
Azalia Mirhoseini @Azaliamirh
14K Followers 519 Following Asst. Prof. of CS at Stanford, Google DeepMind. Prev: Anthropic, Google Brain. Co-Creator of MoEs, AlphaChip, Test Time Scaling Laws.
henry castillo @henrycstllo
79 Followers 613 Following phd @tamu. prev: swe @stripe, bs @utaustin. i want to mechanistically understand models through the lens of training dynamics. 🇵🇪🏳️🌈
Cameron Jones @camrobjones
1K Followers 779 Following Assistant Professor in Psychology at Stony Brook University. I’m interested in how people interact with LLMs and they impact they might have on our psychology.
Nick Haber @nickhaber
890 Followers 254 Following Interactively learning AI, cognitive models, learning tools. Assistant Professor at Stanford.
Arian Hosseini @arianTBD
2K Followers 324 Following Research Scientist @GoogleDeepMind - LLM reasoning and alignment - prev: @Google @MSFTResearch
gabriel @GabrielPeterss4
36K Followers 489 Following research sora at @OpenAI, previously at midjourney, swedish high school dropout
Shengjia Zhao @shengjia_zhao
52K Followers 230 Following Chief Scientist @ Meta MSL. Formerly MTS @ OpenAI, PhD @ Stanford. I train models. All opinions my own.
Chen Sun 🤖🧠🇨... @ChenSun92
2K Followers 398 Following Research Scientist @ Google DeepMind Building memory & open-ended AI ex-neuroscientist ex-IMO team Canada Views are mine alone not GDM's.
Norman Mu @TheNormanMu
2K Followers 803 Following
Xinyun Chen @xinyun_chen_
7K Followers 1K Following Research Scientist @Meta MSL. Prev. @GoogleDeepMind. PhD @Berkeley_EECS.
Roberta Raileanu @robertarail
9K Followers 2K Following Senior Staff Research Scientist @GoogleDeepMind & Honorary Lecturer @UCL. ex @Meta|@MSFTResearch|@NYU|@Princeton. Llama-3, Toolformer, Rainbow Teaming, MLGym.