Goto

Collaborating Authors

 sally ride


Unsupervised decoding of encoded reasoning using language model interpretability

Fang, Ching, Marks, Samuel

arXiv.org Artificial Intelligence

As large language models become increasingly capable, there is growing concern that they may develop reasoning processes that are encoded or hidden from human oversight. To investigate whether current interpretability techniques can penetrate such encoded reasoning, we construct a controlled testbed by fine-tuning a reasoning model (DeepSeek-R1-Distill-Llama-70B) to perform chain-of-thought reasoning in ROT-13 encryption while maintaining intelligible English outputs. We evaluate mechanistic interpretability methods--in particular, logit lens analysis--on their ability to decode the model's hidden reasoning process using only internal activations. We show that logit lens can effectively translate encoded reasoning, with accuracy peaking in intermediate-to-late layers. Finally, we develop a fully unsupervised decoding pipeline that combines logit lens with automated paraphrasing, achieving substantial accuracy in reconstructing complete reasoning transcripts from internal model representations. These findings suggest that current mechanistic interpretability techniques may be more robust to simple forms of encoded reasoning than previously understood. Our work provides an initial framework for evaluating interpretability methods against models that reason in non-human-readable formats, contributing to the broader challenge of maintaining oversight over increasingly capable AI systems.


NASA's trailblazing generation

Popular Science

NASA's Class of 1978, 'represent the most competent, talented, and experienced people available to us today.' The first six women in newly issued, incompletely adorned astronaut jumpsuits, 1978: (front, left to right) Sally Ride, Rhea Seddon; (rear) Kathy Sullivan, Shannon Lucid, Anna Fisher, Judy Resnik. Breakthroughs, discoveries, and DIY tips sent every weekday. Reprinted with permission of the publisher, Smithsonian Books. Out October 28 and available wherever books are sold. Members of the media peppered administrator Robert Frosch with questions and sought assurances about the selection process, the number of women and people of color, and the number of military and civilian pilots selected. Chris Kraft, director of the Johnson Space Center, fielded questions and explained the experience-based filters and rating process for competitive selection. He was satisfied that the men and women selected "represent the most competent, talented, and experienced people available to us today." The main press conference to introduce the new astronaut candidates to the public occurred on January 31.


On this day in history, June 18, 1983, astronaut Sally Ride becomes first American woman in space

FOX News

Astronaut Sally Ride became the first American woman in space on this day in history, June 18, 1983. Born on May 26, 1951, in Los Angeles, Ride earned bachelor's degrees in English and physics from Stanford University in California before staying at Stanford and earning a PhD in physics in 1978. Shortly before earning her doctorate, Ride saw an ad for a newspaper that piqued her interest. NASA was recruiting for astronauts -- and, for the first time, the agency would include women in its astronaut class. "Over 8,000 men and women applied to the space program that year. Of the 35 individuals accepted, six were women, and I was one of them. This was in January 1978," said Ride in quotes listed on a tribute page on NASA's website.


Sally Ride's NASA Career Was Even More Interesting Than You Thought

Forbes - Tech

Sally Ride was the first astronaut to operate the robotic Canadarm. We know Sally Ride, born on this day in 1951, as the first American woman in space (and since her death in 2012, we've known her as the only openly LGBT American astronaut so far) but she did much more than just show up for the ride, whether on her first Space Shuttle flight or during her 34-year career as an astronaut and physicist. It's easy for the basic fact of Ride's identity to eclipse her actual work. Her 1983 Challenger flight was groundbreaking, but in many ways the rest of her career is far more interesting; Ride herself certainly thought so, based on her comments in several interviews. From designing Canadarm 1 to investigating the loss of two Space Shuttles, and from X-rays in the interstellar medium to the physics of lasers, here's a look at what the first American woman in space actually did.