Gutenberg @gutenbergpbc
https://t.co/7izaqrxp6d Joined November 2025-
Tweets3
-
Followers17
-
Following25
-
Likes31
What features constitute AI-generated text? We used our data-centric interp SDK on @pangram's public dataset and found some interesting features
New paper! We used Sparse Autoencoder (SAE) embeddings to understand how agent behavior actually changes during post training. We analyzed thousands of rollouts in a Diplomacy environment and uncovered generalized reward hacking, weird roleplay, and the root cause of an unsuccessful training run. All of which our LLM baseline couldn't find.
I’m excited to share a bit about what I’ve been working on in data-centric interpretability. It’s a paradigm with great potential, and it’s ready for application. Our production system has already been tested on real customers’ data to improve their engagement metrics. But we can’t share that, so we’ve made a synthetic dataset to demonstrate. Basically, pretend you’re a marketing firm trying to generate messages that engage users, and they have this hidden preference in the data. LLMs and embeddings fail to catch it, but our system can. Data-centric interp is efficient and good at discovering unknown unknowns. We’ve demonstrated it works on short texts. But what’s really exciting is to extend this to longer contexts and larger datasets. Monitoring agent traces is costly. Training datasets are too massive to even try. But data-centric interp gives you rich representations of data, which can systematically uncover insights, anomalies, and other qualities over large datasets. We can help find these hidden features in your data that will drive your metrics. I’ll be at NeurIPS and am always down to chat data-centric interp
George Ingebretsen @georgeing
898 Followers 839 Following AI Village (@aidigest_); prev @CAIS, @CHAI_Berkeley
Elvis Sikora @SikoraElvis
4 Followers 8 Following Independent researcher. Forecasting, memory systems, AI swarms.
Qiang @qiangw26
8 Followers 561 Following
Ryan Conti @ContiThinking
65 Followers 780 Following thinking and learning about anything 🙂 productionizing useful interp @Gutenbergpbc Indigenous language work @MILA | Chikashshįyaakni' Math & CS @ MIT
Denis Lim @denisfedelim
2 Followers 84 Following Visiting Researcher at @KAIST_AI Research Fellow at SPAR (Metacognition, RL) Prev. Scale AI, Two Sigma Metacognition, Mech. Interp, AI Alignment & Safety
Luke Versweyveld @LukeVerswey
1K Followers 400 Following hand off your back-office to a tech-enabled team that works like a single, impossibly capable employee
Ivar Frisch @FrischIvar
138 Followers 801 Following philosophy of tech + Antikythera research fellow. LLM psychology research at @acsresearchorg
Ashay Srivastava @AshaySriva82485
12 Followers 61 Following
Lysander Mawby @LysanderMawby
16 Followers 37 Following
Quentin Feuillade--Mo... @LLMPsychologist
324 Followers 135 Following Seldon Lab Batch #2. In SF building the future of programming
Finn Metz @finnlmetz
213 Followers 480 Following It doesn't help if you have the tallest ladder when you're going to the moon. co-founder @seldonai Views are 81.5% my own
John Y 🔸 @yanjo115
444 Followers 2K Following Gutenberg. ex-anthropic, meta 🔸 10% Pledge with @givingwhatwecan
will depue @willdepue
79K Followers 3K Following dei ex machina prev @ openai (sora 1 & 2, posttraining o3/4o, pretraining moonshots)
Metaplanet @metaplanetfund
511 Followers 31 Following Metaplanet is an evergreen investment firm that backs missionaries who build safe technologies for humanity
Logan Graham @logangraham
25K Followers 9K Following Head of the Frontier Red Team @anthropicai. 🌎 Make things radically good.
Tim Hua 🇺🇦 @Tim_Hua_
3K Followers 2K Following AI safety, Econ, new liberalism, math, and a bit of art history (as a treat) Member of Technical Staff @METR_Evals. Previously on Walmart’s economics team.
Ryan Kidd @ryan_kidd44
4K Followers 2K Following Building the AI safety & security field @MATSprogram
Jason Wolfe @w01fe
6K Followers 840 Following alignment and the model spec @OpenAI (opinions are my own)
Behnam Neyshabur @bneyshabur
47K Followers 1K Following Co-Founder & CEO @mirendil 💼 Past: co-led Discovery team @AnthropicAI & Blueshift team @GoogleDeepMind 🎒Traveling & Backpacking
Neev Parikh @neev_parikh
2K Followers 2K Following are you ready for the intelligence explosion anon? @AnthropicAI. prev @METR_Evals, @Stripe, opinions my own. part time angel.
Vaidehi Agarwalla @vaidehiagrwalla
2K Followers 758 Following I like updating (my beliefs). 🇸🇬🇮🇳 🔸10% pledger
George Ingebretsen @georgeing
898 Followers 839 Following AI Village (@aidigest_); prev @CAIS, @CHAI_Berkeley
Peter Wildeford🇺�... @peterwildeford
39K Followers 410 Following Head of Policy @AIPolicyNetwork. Ohioan. World champion forecaster. Former data scientist. Protecting liberty and prosperity in the age of superintelligence.
Alexander Panfilov @kotekjedi_ml
10K Followers 471 Following PhD @ELLISInst_Tue & @MPI_IS | past: @MATSprogram 9.0 fellow doing AI Safety & Adversarial ML
Severin Field 🦅 @sevdeawesome
2K Followers 678 Following i quit boogie boarding to solve 🇺🇸 AI policy
Rob Miles @robertskmiles
67K Followers 871 Following Explaining AI Alignment to anyone who'll stand still for long enough, on YouTube and Discord. Music, movies, microcode, and high-speed pizza delivery
SemiAnalysis @SemiAnalysis_
179K Followers 38 Following
Naomi Saphra @nsaphra
11K Followers 1K Following Waiting on a robot body. All opinions are universal and held by both employers and family. Now a dedicated grok hate account. (I post more elsewhere.)
NeurIPS Conference @NeurIPSConf
167K Followers 41 Following Sydney Dec 6-12, 26, Paris and Atlanta. Tweets to this account are not monitored. Please send feedback to [email protected].
Miles Brundage @Miles_Brundage
79K Followers 14K Following AI policy researcher, @lfschiavo wife guy, Executive Director of @AVERIorg, Substacker, fan of animals and sci-fi, views my own
levent @__alpoge__
53K Followers 230 Following idiot. cuda og, harvard val, morgan prize, society of fellows, 1 hilbert problem so far, creating friendly, SAFE, delightful, supergenius ..things @anthropicai
Séb Krier @sebkrier
31K Followers 8K Following 🪼 AGI jester | purveyor of machine funk, dimensional glider, deep ArXiv dweller, uncertain interstellar fugitive, views my own & not employer's 🛸
Demis Hassabis @demishassabis
2.0M Followers 179 Following Nobel Laureate. Co-Founder & Chair @GoogleDeepMind; Chief Scientist of Alphabet. Founder & CEO @IsomorphicLabs. Trying to understand the nature of reality...
John Y 🔸 @yanjo115
444 Followers 2K Following Gutenberg. ex-anthropic, meta 🔸 10% Pledge with @givingwhatwecanYou might like
