Sonia Joseph @soniajoseph_
AI researcher. Getting PhD @Mila_Quebec, prev @Princeton. Multimodal interpretability + alignment. San Francisco Joined August 2015-
Tweets516
-
Followers4K
-
Following540
-
Likes3K
Tbh contouring your face is ridiculously harder than coding a transformer from scratch. If make up artists applied this level of attention to detail to neural nets, we’d have a renaissance of architectures— it’s all just shape rotation.
🚀 Thrilled to introduce Recall to Imagine (R2I), the 1st model-based RL approach integrating SSMs to excel in memory-intensive domains. Not just setting new SOTA, but achieving superhuman results in complex memory tasks, while efficiently operating across diverse domains. 1/
🚀 Thrilled to introduce Recall to Imagine (R2I), the 1st model-based RL approach integrating SSMs to excel in memory-intensive domains. Not just setting new SOTA, but achieving superhuman results in complex memory tasks, while efficiently operating across diverse domains. 1/
I'll be supervising Mechanistic Interpretability at this summer's MATS program-fill in this short airtable form by Sunday (24th March)! airtable.com/appPxJ0QMqR7TE… From my mentee and (co-)mentor experience, it's a great program to do important work understanding how AIs work!
Just published a set of Sparse AutoEncoders GPT2-Small residual stream all layers. lesswrong.com/posts/f9EgfLSu…. Thanks @NeelNanda5 and @ArthurConmy for the support!
Are you excited about @ch402-style mechanistic interpretability research? I'm looking for scholars to mentor via MATS - apply by April 12! I'm very impressed by the great work from past scholars, and enjoy mentoring promising mech interp talent. I'm excited for my next cohort!
The Prisma multi modal mech interp discord is shaping up to be the most fascinating blend of AI alignment people and artsy designer stable diffusion esque people. The subculture I never knew I needed. discord.com/invite/2U2N8Qm…
Ethan Caballero is bu.. @ethanCaballero
8K Followers 2K Following ML PhD student @Mila_Quebec ; previously @GoogleDeepMind
Irina Rish @irinarish
9K Followers 994 Following prof UdeM/Mila; Canada Excellence Research Chair; AAI Lab head https://t.co/UzlrC7ZrGF; INCITE project PI https://t.co/0rV7szd7rH; CSO https://t.co/XDhj6MEtUj
Miles Brundage @Miles_Brundage
43K Followers 10K Following Policy research at @openai. I mostly tweet about AI, animals, and sci-fi. He/him. Views my own.
EigenGender @EigenGender
6K Followers 661 Following all my posts are shitposts that simultaneously reveal the true nature of reality. large language models; kinda EA; 🏳️⚧️
Noam Brown @polynoamial
34K Followers 612 Following Researching reasoning @OpenAI | Co-created Libratus/Pluribus, the first superhuman no-limit poker AIs | Co-created CICERO | PhD from @SCSatCMU
Hattie Zhou ✈️ IC.. @oh_that_hat
5K Followers 765 Following Finding \hat{y} Give me anonymous feedback: https://t.co/7aBNrpaCnA
David Krueger @DavidSKrueger
13K Followers 4K Following Cambridge faculty - AI alignment, deep learning, and existential safety. Formerly Mila, FHI, DeepMind, ElementAI, AISI.
Patrick Mineault @patrickmineault
19K Followers 3K Following Neuro AI, vision, Python, open science. Senior ML scientist @ Mila. Previously engineer @ Google, Meta. Updates from https://t.co/d0o7cSLC6o, https://t.co/duh0cFwLyw
Keerthana Gopalakrish.. @keerthanpg
13K Followers 830 Following Building Embodied AGI. Research @DeepMind. Opinions my own.
Mo Bavarian @mobav0
11K Followers 918 Following Research Scientist, working on optimization and architecture of LLMs at OpenAI. Math ❤️. Prev SWE Rubrik, PhD MIT.
Sara 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.
Stella Biderman @BlancheMinerva
15K Followers 749 Following Open source LLMs and interpretability research at @BoozAllen and @AiEleuther. My employers disown my tweets. She/her
sigfig @sigfig
9K Followers 745 Following i saw a space ship blow up once and immediately decided i wanted to get in one
Stanislas Polu @spolu
14K Followers 606 Following _co-founder+engineer(https://t.co/fCirsLjeo2), _alumni(https://t.co/8jAnpFAkp1, https://t.co/e99AaHzlA0, https://t.co/4jg6knqi2S, https://t.co/kXE6PNf8xH)
Eugene Vinitsky @EugeneVinitsky
13K Followers 2K Following Lets make multi-agent learning easy. Anti-cynic. RS at Apple, Asst. Prof at @nyutandon. He/him. Anonymous feedback: https://t.co/Mmmg7uPm1t
Alex K. Chen... @InquilineKea
3K Followers 3K Following Interesting things happen around me.🦜🦜Meta+zany+altricial distributed computation. My likes are the best likes. Ideal figure to slow world aging rate
ÀI Enthusiast @wisdomvaultai
34 Followers 349 Following
Dylan Itzikowitz @ditzikow
3K Followers 2K Following 🇨🇦 in 🇺🇸 Investing in early stage founders @Southpkcommons | DM's open | Ex @Expa @foundersfactory 🇬🇧 | Founder @theforwardmove | “Habitual dot connector”
Andre Charoo @acharoo
4K Followers 2K Following GP @MapleVCFund. Venture Partner @Inovia. Co-Chair @theC100. First 25 @Uber and @Hired_HQ.
Imad Khwaja @flyingblackswan
167 Followers 2K Following SaaS Growth || SEO Marketing Agency || Entrepreneur
Aghyad Deeb @aghyadd98
2 Followers 27 Following
Athina Sotiropoulou @AthenaSotirop
45 Followers 371 Following MSc student Computational Science & Engineering @ETH_en
Marco Marinucci @MMarinucciVC
629 Followers 787 Following Co-Founder / General Partner of Essentia Ventures, Co-Founder and ex MP HELLA Ventures #Investor #VC #GenerativeAI #Mobility #EV #SiliconValley
Vivek Reddy @vivekviga
233 Followers 2K Following Tweets about ML + Real-time rendering . Founder/CEO @viga_et . TD on Lion King 2019. @upenn alum
Pensé FFun @inftyCategory
93 Followers 6K Following
Joseph Melkonian @joemelko
72 Followers 152 Following building @womplabsai | @aigrant batch 3 | @join_ef W23 | 0.00000036% of the way to raising $7 trillion
Jason Carman @jasonjoyride
21K Followers 2K Following Launching satellites @astranis & weekly startup documentaries via S³
Matt Doll @MatthewRDoll
20 Followers 34 Following
Guneet Singh Kohli @guneetsk99
468 Followers 3K Following AI Engineer @ GreyOrange, Building Indian LLMs with Odia GenAI Independent Researcher working on variety of random problems.
John Wayne @WeiJian47770542
13 Followers 181 Following
Harsh Desai @dreamerharsh
1 Followers 3K Following
David @writerdaviddyck
922 Followers 4K Following Writer, Reader, Yarn Spinner, Truth-Sayer, Highly Disquieted Denizen of Humanity
lucy wang @lucyywang22
536 Followers 336 Following
AI World News🤖💕 @TheWorldNews
56K Followers 9K Following The first rule of AI club is, We Don't Talk About AI Club! The second rule of AI Club is......
Alex Winters @axwinters
6 Followers 184 Following Getting stats to mean something, then getting stats to think something.
Jeanne @suzanne_ellas_
0 Followers 29 Following
technotechies @techno_techies
71 Followers 772 Following
S. Iqbal @S_Iqbal90
28 Followers 1K Following
ີີີີີີ່.. @gpu_poor
108 Followers 509 Following
divya reddy vemunuri @vemunuri_divya
0 Followers 70 Following
Antonin Poché @Antonin_Poche
23 Followers 50 Following Research engineer in XAI at IRT Saint Exupéry in https://t.co/RhQcHCalSF and https://t.co/pBG7yb0Hnw projects. In the developing team of Xplique.
Tuomas Oikarinen @tuomasoi
62 Followers 147 Following Developing scalable ways to understand neural networks. PhD student at UCSD. https://t.co/aiLkcm9OID
Joseph Bendana @joseph_bendana
101 Followers 124 Following I used to think a lot about thinking. I still do, but I used to too.
danny @danny_nkjg
70 Followers 137 Following ZFellow | Physics | Gorgeous Math | AI Generally hate short form text content (ie tweeting); if you want a textually psychedelic experience, check out my blog
Dr. Kunal Suri @kunalsuri
491 Followers 2K Following Dreamer, Computer Scientist, Engineer, and in an eternal dilemma, whatever that means! I help to create technologies to make the world a better place! ✌️
Simon Spavound @SimonSpavound
169 Followers 551 Following Head of Data Science, US Ops @Peak_HQ, Forecasting, Economics and #rstats enthusiast. PhD Economics from @LancasterUni. Views my own.
Alex Booth @geoesoteric
46 Followers 72 Following life flowing and seeking its own dynamic balance in countless seen and unseen ways
Christoph Schuringa @chrisschuringa
6K Followers 4K Following writing A Social History of Analytic Philosophy (Verso); Karl Marx and the Actualization of Philosophy (Cambridge University Press)
Nalin Gupta @nalingupta01
1K Followers 4K Following YC founder with 1 exit in self-driving cars and Forbes 30U30 alum. Building Wingman. Previously co-founder @Auro_ai (YC, acq'd by @ridecell)
Yang Fan 范阳 @Yang_Supertramp
884 Followers 5K Following Will visit the US. in 2024. Figuring out AI x Bio intelligence, building sth new. ❤️ 🤖️🧬 🧠 🧘 🌍 Member of The Explorers Club.
Yehaoyan_bbb @yan_bbbb
26 Followers 156 Following
jack morris @jxmnop
11K Followers 772 Following getting my phd in nlp @cornell_tech 🚠 // academic optimist // tweeting from the snack aisle at trader joes
Parul Pandey @pandeyparul
7K Followers 1K Following Author | Principal Data Scientist @h2oai | @kaggle Grandmaster(Notebooks) | Mom
lidya @lydiafarnham
5 Followers 88 Following
Charlotte Pouw @pouwcharlotte
65 Followers 76 Following PhD student @illc_amsterdam, working on the interpretability of deep learning models for speech, text and music.
Giulia Rambelli @g_rambelli
391 Followers 805 Following 🎓 Postdoctoral researcher for @Abstraction_ERC project at @Unibo 🤿 Freediver
Ethan Caballero is bu.. @ethanCaballero
8K Followers 2K Following ML PhD student @Mila_Quebec ; previously @GoogleDeepMind
Irina Rish @irinarish
9K Followers 994 Following prof UdeM/Mila; Canada Excellence Research Chair; AAI Lab head https://t.co/UzlrC7ZrGF; INCITE project PI https://t.co/0rV7szd7rH; CSO https://t.co/XDhj6MEtUj
Miles Brundage @Miles_Brundage
43K Followers 10K Following Policy research at @openai. I mostly tweet about AI, animals, and sci-fi. He/him. Views my own.
Neel Nanda @NeelNanda5
14K Followers 89 Following Mechanistic Interpretability lead @DeepMind. Formerly @AnthropicAI, independent. In this to reduce AI X-risk. Neural networks can be understood, let's go do it!
Percy Liang @percyliang
50K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | Pianist
Noam Brown @polynoamial
34K Followers 612 Following Researching reasoning @OpenAI | Co-created Libratus/Pluribus, the first superhuman no-limit poker AIs | Co-created CICERO | PhD from @SCSatCMU
Hattie Zhou ✈️ IC.. @oh_that_hat
5K Followers 765 Following Finding \hat{y} Give me anonymous feedback: https://t.co/7aBNrpaCnA
David Krueger @DavidSKrueger
13K Followers 4K Following Cambridge faculty - AI alignment, deep learning, and existential safety. Formerly Mila, FHI, DeepMind, ElementAI, AISI.
Patrick Mineault @patrickmineault
19K Followers 3K Following Neuro AI, vision, Python, open science. Senior ML scientist @ Mila. Previously engineer @ Google, Meta. Updates from https://t.co/d0o7cSLC6o, https://t.co/duh0cFwLyw
Keerthana Gopalakrish.. @keerthanpg
13K Followers 830 Following Building Embodied AGI. Research @DeepMind. Opinions my own.
Mo Bavarian @mobav0
11K Followers 918 Following Research Scientist, working on optimization and architecture of LLMs at OpenAI. Math ❤️. Prev SWE Rubrik, PhD MIT.
Sara 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.
Stella Biderman @BlancheMinerva
15K Followers 749 Following Open source LLMs and interpretability research at @BoozAllen and @AiEleuther. My employers disown my tweets. She/her
sigfig @sigfig
9K Followers 745 Following i saw a space ship blow up once and immediately decided i wanted to get in one
rohan anil @_arohan_
12K Followers 2K Following Principal Engineer, @GoogleDeepMind Gemini. prev PaLM-2. Tinkering with optimization and distributed systems. opinions are my own.
Stanislas Polu @spolu
14K Followers 606 Following _co-founder+engineer(https://t.co/fCirsLjeo2), _alumni(https://t.co/8jAnpFAkp1, https://t.co/e99AaHzlA0, https://t.co/4jg6knqi2S, https://t.co/kXE6PNf8xH)
SMA 🏴☠️ @generic_void
8K Followers 2K Following dark empress. e/acc. shapecelword-rotator. transhumanist. COC @ MASI. {Research} ⊃ {Post-Scarcity Econ, Game/Info/Network Theory, ML/DL & Econometrics, ..., N}
Dylan Itzikowitz @ditzikow
3K Followers 2K Following 🇨🇦 in 🇺🇸 Investing in early stage founders @Southpkcommons | DM's open | Ex @Expa @foundersfactory 🇬🇧 | Founder @theforwardmove | “Habitual dot connector”
Evan Tana | evantana... @evantana
3K Followers 540 Following Early stage partner to founders @southpkcommons, @ScriptCapital. Formerly @dropbox, @GreylockVC, @shopkick, Loopt (YC S05). SF local and lover of life's quirks.
Joseph Melkonian @joemelko
72 Followers 152 Following building @womplabsai | @aigrant batch 3 | @join_ef W23 | 0.00000036% of the way to raising $7 trillion
Jason Carman @jasonjoyride
21K Followers 2K Following Launching satellites @astranis & weekly startup documentaries via S³
Andre Charoo @acharoo
4K Followers 2K Following GP @MapleVCFund. Venture Partner @Inovia. Co-Chair @theC100. First 25 @Uber and @Hired_HQ.
Marc Andreessen 🇺�.. @pmarca
1.4M Followers 24K Following Techno-optimist. E/acc. Technology brother. Move Fast and Make Things. p(Doom) = 0; p(“1984”) = not 0.
Paul Graham @paulg
1.9M Followers 772 Following
Naveen 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.
Dhruv Batra @DhruvBatraDB
14K Followers 327 Following Senior Director (FAIR @MetaAI). Professor (@GeorgiaTech). Co-founded CaliperAI. Researcher in AI. @CarnegieMellon alum.
Luke Mercado @LukeMercado11
61 Followers 199 Following Fmr. Head of Engineering @ Daybreak Health, Software by day, economics, politics and military history by night.
Sanjeev Arora @prfsanjeevarora
21K Followers 32 Following Director, @PrincetonPLI and Professor @PrincetonCS. Seeks math/conceptual understanding of deep learning and large AI models.
Alexa R. Tartaglini @ARTartaglini
437 Followers 483 Following Research scientist @ the Human & Machine Learning Lab, NYU // Interested in representational alignment, interpretability, & more esoteric things
Zeyuan Allen-Zhu @ZeyuanAllenZhu
8K Followers 274 Following physics of language models @ Meta / FAIR IOI - USACO - MCM - ACM/ICPC - Codejam Tsinghua - MIT - Princeton/IAS - MSR - FAIR
Founders Fund @foundersfund
111K Followers 170 Following technology, science, venture capital // what does the future look like?
Hyperstition, Inc @hyperstition_x
4K Followers 1 Following The future is forged by the hammer of cognition against the anvil of creation.
George A. Michel @aGeorgeMichel
39 Followers 58 Following Tinkerer and aberrant philosopher. Former Toyota/Brex/Flux. Founder at Nomad Science LLC. Looking to make some cool P2P stuff for my hometown of Sacramento, CA.
Dwarkesh Patel @dwarkesh_sp
55K Followers 702 Following Being pretrained Host of Dwarkesh Podcast https://t.co/3SXlu7fy6N https://t.co/rEhnfYywXY https://t.co/hQfIWdM1Un
Tuomas Oikarinen @tuomasoi
62 Followers 147 Following Developing scalable ways to understand neural networks. PhD student at UCSD. https://t.co/aiLkcm9OID
Marco Marinucci @MMarinucciVC
629 Followers 787 Following Co-Founder / General Partner of Essentia Ventures, Co-Founder and ex MP HELLA Ventures #Investor #VC #GenerativeAI #Mobility #EV #SiliconValley
Samuel Marks @saprmarks
720 Followers 79 Following Postdoc studying interpretability for AI safety under @davidbau. PhD in math from @harvard. Previously director of technical programs at https://t.co/FxRv4QgERO.
Christoph Schuringa @chrisschuringa
6K Followers 4K Following writing A Social History of Analytic Philosophy (Verso); Karl Marx and the Actualization of Philosophy (Cambridge University Press)
Ryan Lowe @ryan_t_lowe
5K Followers 359 Following what is the place from which we are creating? ❤️✨🤠❤️
jack morris @jxmnop
11K Followers 772 Following getting my phd in nlp @cornell_tech 🚠 // academic optimist // tweeting from the snack aisle at trader joes
Minh Nhat Nguyen @menhguin
490 Followers 2K Following AI Safety research+climate protests. Owner of @AIHubCentral. Product @hume_ai. Don't do the deferred life plan.
Sarath Chandar @apsarathchandar
4K Followers 423 Following Assistant Professor @polymtl and @Mila_Quebec; Canada CIFAR AI Chair; Machine Learning Researcher. Pro-bono office hours: https://t.co/tK69DKRf9N?amp=1
Lucy Li @lucy3_li
4K Followers 2K Following @UCBerkeley PhD student + @allen_ai. Human-centered #NLProc, computational social science, AI fairness. she/her. https://t.co/rtSSUhWQnL
Astasia Myers @AstasiaMyers
4K Followers 1K Following General Partner at @Felicis. Prev. @Redpoint @QuietCapital & @cisco_invests. Blog: https://t.co/fZRzRA4Yi2
Jeff Dean (@🏡) @JeffDean
297K Followers 6K Following Chief Scientist, Google DeepMind and Google Research. Co-designer/implementor of things like @TensorFlow, MapReduce, Bigtable, Spanner, Gemini .. (he/him)
Sholto Douglas @_sholtodouglas
15K Followers 862 Following Scaling Gemini @Deepmind - working towards intelligence too cheap to meter
Alexander Chen @alexanderchen
8K Followers 1K Following Creative Director at Google Creative Lab, working on AI. Opinions are my own. https://t.co/bSOAeDObmz
Andrei Mircea @mirandrom
53 Followers 301 Following PhD student @Mila_Quebec ⊗ mechanistic interpretability + systematic generalization + LLMs for science ⊗ https://t.co/xg8aE8CoWv
Ying Shan @yshan2u
1K Followers 602 Following Distinguished Scientist @TencentGlobal, Founder of PCG ARC Lab, Director of AI Lab Visual Computing. Formerly @Microsoft, @MSFTResearch. Views are my own.
Omar Sanseviero @osanseviero
32K Followers 2K Following Chief Llama Officer @huggingface 🦙 Founder @AI_Learners. Xoogler (SWE @Google Assistant, 20% PM TF Graphics). 100% Hacker Llama🇵🇪🇲🇽
Connor Kissane @Connor_Kissane
49 Followers 42 Following Mechanistic Interpretability research / software engineering
Juan A. Rodríguez @joanrod_ai
131 Followers 353 Following PhD Student at @Mila_Quebec and @etsmtl and researching at @ServiceNowRSRCH in Montreal. Previously at UPF and UAB-CVC. Working on Multimodal Generative Models.
Dane Malenfant @dvnxmvl_hdf5
21 Followers 76 Following MSc. Computer Science @Mila_Quebec & @mcgillu in the LiNC lab | Currently distracted with multi-agent RL and neuroAI | Restless | Ēka ē-akimiht
Romeo Valentin @romeo_valentin_
25 Followers 282 Following PhD @SISLaboratory. Real-world ML safety. Open source open science. @JuliaLanguage believer. Meme connoisseur. Ex @ETH_en, @RWTH, @CMUEngineering.
Jacob Steinhardt @JacobSteinhardt
7K Followers 67 Following Assistant Professor of Statistics, UC Berkeley
Xiaohua Zhai @XiaohuaZhai
3K Followers 208 Following Senior Staff Researcher @GoogleDeepMind team in Zürich
@soniajoseph_ Yes! It's fucking fantastic and so few people I know have read it. You have to read Exordia by the same guy, it's about aliens, first contact, and how the sketches of ethical systems are filled in by the kind of people who choose them.
@soniajoseph_ If you have a research project that needs the training data, why not use the CLIP models actually trained on LAION instead of OpenAI's?
@soniajoseph_ I guess it depends more on the jurisdiction you fall under than on your personal opinion.
If I could go back in time, I would have worked on NeurIPS papers in high school.
@zamresearch Maybe, but I'm reluctant to do that sort of thing. Online controversies are mostly driven by idiots, so if you write essays in response to online controversies, you're letting your agenda be determined by idiots.
I have been working on vision+language models (VLMs) for a decade. And every few years, this community re-discovers the same lesson -- that on difficult tasks, VLMs regress to being nearly blind! Visual content provides minor improvement to a VLM over an LLM, even when these…
Today we’re releasing OpenEQA — the Open-Vocabulary Embodied Question Answering Benchmark. It measures an AI agent’s understanding of physical environments by probing it with open vocabulary questions like “Where did I leave my badge?” More details ➡️ go.fb.me/7vq6hm…
Really excited about the launch of this research initiative. Hiring Research Scientists now. Research Software Engineers and postdocs over next few months. 300 H100 GPUs. Multidisciplinary teams. Princeton helps keep AI expertise in the open sphere. More: pli.princeton.edu
“The dramatic rise of AI capabilities…is a watershed event for humanity…It is also sure to transform research and teaching in every academic discipline.” – @prfsanjeevarora, director of the new @Princeton Language and Intelligence initiative. For more: pli.princeton.edu
Big congratulations to Avi Wigderson of IAS Princeton for winning the Turing Award in CS. Truly an all-time great in theoretical computer science and discrete math. Also one of the nicest human beings I know --friend and mentor to so many (including me) tinyurl.com/fz5vxxaf
@soniajoseph_ Really? I am not sure about this - I find applying makeup orders of magnitude easier (and more enjoyable) than any of that more intellectual stuff :) maybe I chose a wrong profession :)
POV: Me at an social event.
Tbh contouring your face is ridiculously harder than coding a transformer from scratch. If make up artists applied this level of attention to detail to neural nets, we’d have a renaissance of architectures— it’s all just shape rotation.
@soniajoseph_ I have a promising alternative to neural networks and GOFAI. Let me know if you want to read the draft of my book or a paper about generalization.
anyone know what's the best image to text / image captioning model I can run locally with <=20GB vram?
Our 12 scaling laws (for LLM knowledge capacity) are out: arxiv.org/abs/2404.05405. Took me 4mos to submit 50,000 jobs; took Meta 1mo for legal review; FAIR sponsored 4,200,000 GPU hrs. Hope this is a new direction to study scaling laws + help practitioners make informed decisions
"Foundation models" companies that require up-front investment for GPUs are more like like airline companies than software. Large capex, then compete against 50 other companies that all bought the same model of airplane as you.
We are coming out of stealth today and announcing our $25m Seed Round led by @foundersfund and @8vc with participation from @SimoneSyed, @Quantonation, @hankejh, @JTLonsdale, @C_Angermayer, @mattturck We are creating the future of Material Science. Here's the deck 🧵:
Can we understand & edit unanticipated mechanisms in LMs? We introduce sparse feature circuits, & use them to explain LM behaviors, discover & fix LM bugs, & build an automated interpretability pipeline! Preprint w/ @can_rager, @ericjmichaud_, @boknilev, @davidbau, @amuuueller
Trends for United States
You might like
