Sabitlenmiş Tweet
Eric Todd
158 posts
Eric Todd
@ericwtodd
Computer Science PhD Student at Northeastern University
Boston, MA Katılım Aralık 2014
473 Takip Edilen490 Takipçiler
Eric Todd retweetledi
The most popular way to interpret AI is missing the bigger picture.
Models think in curved shapes. But sparse autoencoders (SAEs) work with straight lines.
Can they still capture models’ curved neural geometry? Yes, but not how you might think! (1/7)
Goodfire@GoodfireAI
Neural networks might speak English, but they think in shapes. Understanding their rich *neural geometry* is key to understanding how they work – and to debugging and controlling them with precision. Starting today, we’re releasing a series of posts on this research agenda. 🧵
English
Eric Todd retweetledi
Neural networks have beautiful feature geometry, but do they have mechanisms that actually interface with those structures?
At @GoodfireAI this spring, we discovered one: a re-usable addition mechanism that reads/writes to Fourier features from prior work. 🧵
Goodfire@GoodfireAI
Neural networks do math by rotating shapes. We found a shape-rotating calculator hidden inside an LLM – and it’s used for more than just math! (1/6)
English
Eric Todd retweetledi
Interpretability provides a toolset for understanding how and why LMs behave in certain ways. This survey proposes a perspective on interpretability research grounded in causal mediation analysis: doi.org/10.1162/COLI.a… #NLProc #CLJournal @SunJiuding @ericwtodd
English
Eric Todd retweetledi
1/ (New paper!)
If swapping the gender in an input prompt makes the AI model give a different answer it means that it has to have a gender bias, right? Wrong.
🧵on counterfactual prompting for LLM evals:
Paper: arxiv.org/abs/2605.01048
English
Eric Todd retweetledi
The Teleport Contest is open.
Port NetHack 5.0 from C to JavaScript, bit-exactly. Same screen, every keystroke. Any approach: LLM agents, hand-coded, transpiler, hybrid. Live leaderboard, two phases through December.
mazesofmenace.ai/announcement
English
Eric Todd retweetledi
NetHack is one of the most complex and longest-lived open source programs ever written, and after 46 years, v5.0 shipped today.
nethack.org/common/index.h…
And ... it is a VERY cool large codebase to work with in the LLM era.
English
Eric Todd retweetledi
I'll be attending #ICLR2026 next week to present my work on In-Context Algebra! My poster will be on Fri, April 24 at 3:15-5:45PM at Pavilion 4 P4-#4011. If you're around, stop by and say hello! My DMs are open if you want to connect or meet up in Rio!
Eric Todd@ericwtodd
Can you solve this algebra puzzle? 🧩 cb=c, ac=b, ab=? A small transformer can learn to solve problems like this! And since the letters don't have inherent meaning, this lets us study how context alone imparts meaning. Here's what we found:🧵⬇️
English
Eric Todd retweetledi
2026 is a whirlwind year for AI.
Underlying it all: the greatest scientific mystery of our age. How does a neural network think?
I talked w @oliver_whang22 in NYTimes Magazine, on how AI interpretability is a tangle of structure waiting to be unraveled:
nytimes.com/2026/04/15/mag…
English
Eric Todd retweetledi
Excited to be attending #ICLR in person this year! I’ll be presenting 3 works across the main conference and workshops. If you’re around, please stop by, say hi, and feel free to reach out if you’d like to connect!
English
@trajektoriePL Sounds very similar to this talk from NeurIPS this past year at the mech interp workshop: davidbau.com/archives/2025/…
English
Terence Tao proposes what he calls a "Copernican view of intelligence".
Instead of buying into the common, one-dimensional narrative that artificial intelligence will simply evolve from "subhuman" to "superhuman" and ultimately make humanity entirely redundant, Tao urges us to look at the bigger picture.
Much like the Copernican revolution proved the Earth is not the center of the universe, Tao suggests we need to realize that human intelligence isn't the only, or necessarily the highest, form of intellect. Historically, we have treated other forms of storing or creating knowledge—like animals, books, and computers—as secondary. However, we actually exist within a much richer universe of intelligence.
Both human intelligence and computer intelligence possess their own distinct strengths and weaknesses. The true potential lies not in viewing them as direct competitors, but rather in focusing on collaboration. By working together, humans and computers can achieve additional things that neither could accomplish on their own, requiring us to think in much wider terms than just what humans or computers can do alone.
English
Eric Todd retweetledi
New paper: LLMs encode harmful content generation in a distinct, unified mechanism
Using weight pruning, we find that harmful generation depends on a tiny subset of the weights that are shared across harm types and separate from benign capabilities.
🧵
English
Eric Todd retweetledi
Patients ask LLMs medical questions, but how they phrase it matters more than it should.
Our new preprint explores how different phrasings of patient health questions can lead to inconsistent conclusions, even with the same evidence. [1/6]
Full Paper: arxiv.org/abs/2604.05051
English
Eric Todd retweetledi
If you enjoyed Anthropic's recent emotions paper, check out our pre-print! We find many many similarities:
1) Circular geometry of emotion representations that resembles the "Circumplex Model of Affects" from psychology
2) Steering effects on affective properties of LM outputs -- unlike Anthropic, we steer along the circular manifold (at 0°, 30°, 60°, etc.)
3) Steering effects on other downstream behavior (refusal, sycophancy) -- steering emotion representations can affect refusal/sycophancy rates.
The last one was a bit unexpected - we provide a mechanistic account for why this might happen.
See Lihao's thread below for details!👇
Lihao Sun@1e0sun
💡New paper! Woke up to @AnthropicAI's emotion paper and realized - “wait, that's our finding too.” So we ArXiv'd immediately. We concurrently uncovered a circular geometry of emotions organized by valence and arousal (VA), as well as steering effects on downstream behaviors like refusal and sycophancy. We further provide a mechanistic account for why: refusal and compliance tokens occupy distinct regions in this space. 1/
English
Eric Todd retweetledi
📣 Launching monthly interp puzzles 🧩
Each month: a model trained on a toy task. Your job: reverse-engineer the algorithm it learned.
First puzzle: how does a 1-2L attn-only transformer find the max of a list? Starter Colab included.
Deadline: April 30
puzzles.baulab.info
English
Eric Todd retweetledi
Calling attention to an exciting "deception detection" hackathon we're planning this summer! w @NDIF and @CadenzaLabs.
Recruiting red teams now, blue teams later. Red teams, time is short: proposals due Mar 31. $10K stipend + compute, $15K finals prize.
nnsight.net/blog/2026/03/2…
English
Eric Todd retweetledi
Eric Todd retweetledi
I’ll be presenting our work, “Distilling Diversity and Control in Diffusion Models,” at @wacv_official this Sunday at 11 AM local time.
🔍We uncover the “secret to unlocking diversity” in diffusion models - using **interpretability**!!
DM me if you’d like to connect in Tucson.
Rohit Gandikota@rohitgandikota
Why do distilled diffusion models generate similar-looking images? 🤔 Our Diffusion Target (DT) visualization reveals the secret to diversity. It is the very first time-step! And—there is a simple, training-free way to make them more diverse! Here is how: 🧵👇
English
Eric Todd retweetledi
NNsight 0.6 is out now! We directly address your feedback in our biggest release yet. Pain points included cryptic errors, slow traces, no remote execution of custom code, and limited vLLM support. We tackle all of these and more in this new release.
🧵 Here's what changed:
English