How Does A Blind Model See The Earth? — LessWrong
Sometimes I'm saddened remembering that we've viewed the Earth from space. We can see it all with certainty: there's no northwest passage to search for, no infinite Siberian expanse, and no great uncharted void below the Cape of Good Hope. But, of all these things, I most mourn the loss of incomplete maps. In the earliest renditions of the world, you can see the world not as it is, but as it was to one person in particular. They’re each delightfully egocentric, with the cartographer’s home most often marking the Exact Center Of The Known World. But as you stray further from known routes, details fade, and precise contours give way to educated guesses at the boundaries of the creator's knowledge. It's really an intimate thing. If there's one type of mind I most desperately want that view into, it's that of an AI. So, it's in this spirit that I ask: what does the Earth look like to a large language model? With the following procedure, we'll be able to extract an (imperfect) image of the
x How Does A Blind Model See The Earth? — LessWrong AI Evaluations Language Models (LLMs) Interpretability (ML & AI) World Modeling AI Curated 2025 Top Fifty: 33 % 500 How Does A Blind Model See The Earth? by henry 11th Aug 2025 Linkpost for outsidetext.substack.com 8 min read 42 500 Sometimes I'm saddened remembering that we've viewed the Earth from space. We can see it all with certainty: there's no northwest passage to search for, no infinite Siberian expanse, and no great uncharted void below the Cape of Good Hope. But, of all these things, I most mourn the loss of incomplete maps. In the
saved by
related reading
- How Does A Blind Model See The Earth? - by henryoutsidetext.substack.com
- Simulators — LessWronglesswrong.com
- interpreting GPT: the logit lens — LessWronglesswrong.com
- Gears in understanding — LessWronglesswrong.com
- The World Inside Neural Networksgoodfire.ai
- Language Models, World Models, and Human Model-Buildinglingo.csail.mit.edu
- LLMs are a tool for thoughtwattenberger.com
- A Functional Taxonomy of World Models - Dr. Fei-Fei Lidrfeifei.substack.com
- LLMs and World Models, Part 1 - by Melanie Mitchellaiguide.substack.com
- My picture of the present in AI — LessWronglesswrong.com
- Do large language models understand us? | by Blaise Aguera y Arcas | Mediummedium.com
- llm assistant personas seem increasingly incoherent (some subjective observations) — LessWronglesswrong.com