# A Note on 'Semiotic Physics'  ## Metadata - Author: [[metasemi]] - Full Title: A Note on 'Semiotic Physics' - Category: #articles - Summary: insert summary - My notes: - Summary: GPT generates text by predicting one token at a time, creating many possible story paths like a multiverse. This process is called semiotic physics, where tokens act like particles in a language-based world. Studying these token sequences helps us understand how meaning emerges, but this is different from real-world physics. ## LLM Chats ## NotebookLM ## LLM Audio ## Highlights > GPT, the prototypical [simulator](https://www.lesswrong.com/posts/vJFdjigzmcXMhNTsx/simulators), is often said to "predict the next token" in a sequence of text. This is true as far as it goes, but it only partially describes typical usage, and it misses a dynamic that's essential to GPT's most impressive performances. Usually, we don't simply have GPT predict a single token to follow a given prompt; we have it roll out a continuous passage of text by predicting a token, appending that token to the prompt, predicting *another* token, appending *that*, and so on. ([View Highlight](https://read.readwise.io/read/01k0h1kqb6dmcwtzc6ghq1qeyd)) - Note: good clear description of autoregression and token prediction. > janus (unpublished) says "GPT is analogous to an indeterministic time evolution operator, sampling is analogous to wavefunction collapse, and text generated by GPT is analogous to an Everett branch in an implicit multiverse." ([View Highlight](https://read.readwise.io/read/01k0h21pd84ybk3d77yhz5mzg6)) > Like real-world physics, the simulator's "physics" leads to emergent phenomena of immediate significance to human beings. In real-world physics, these emergent phenomena include stars and snails; in semiotic physics, they're the stories the simulators tell and the [*simulacra*](https://www.lesswrong.com/posts/3BDqZMNSJDBg2oyvW/simulacra-are-things) that populate them. Insofar as these are unprecedented rhymes with human cognition, they merit investigation for their own sake. ([View Highlight](https://read.readwise.io/read/01k0h2c9n6kq9awd3ykf6as4zx)) - Tags: [[semiotic physics]] > It's more illuminating to consider what happens when GPT, or any simulator, is run repeatedly to produce a multi-token forward [trajectory](https://www.lesswrong.com/s/guzvzGnRHzMBWLqKZ/p/TTn6vTcZ3szBctvgb#Simulations_as_dynamical_systems), as in the familiar scenario of generating a text completion in response to a prompt. ([View Highlight](https://read.readwise.io/read/01jzht68zx35d01yvahfj5ej9d)) > In this analogical sense, a simulator such as GPT implements a "physics" whose "elementary particles" are linguistic tokens. When we experience the generated output text as meaningful, the tokens it's composed of are serving as [semiotic signs](https://en.wikipedia.org/wiki/Sign_(semiotics)). Thus we can refer to the simulator's physics-analogue as *semiotic physics*. ([View Highlight](https://read.readwise.io/read/01jzht6vrqv14hq7w8h8y97rvc)) > As linguistically capable creatures, we experience the simulator's outputs as *semantic*. The tokens in the generated trajectory carry meaning, and serve as [semiotic signs](https://en.wikipedia.org/wiki/Sign_(semiotics)). This is why we refer to the simulator's physics-analogue as *semiotic physics*. ([View Highlight](https://read.readwise.io/read/01k0h34awbee2vc9tqfmtbvxn9)) > Semiotic physics represents a [naturalistic](https://www.lesswrong.com/s/evLkoqsbi79AnM5sz) method of exploring the simulator from the output side that contrasts with and complements other (undoubtedly important) approaches such as "[thinking about] exactly what is in the training data", as [Beth Barnes has put it](https://www.lesswrong.com/posts/dYnHLWMXCYdm9xu5j/simulator-framing-and-confusions-about-llms). ([View Highlight](https://read.readwise.io/read/01k0h36nbqr0xqgd6zsyepvwrp)) > It's in this analogical sense that a simulator like GPT implements a "physics" whose "elementary particles" are linguistic tokens. > Like real-world physics, the simulator's "physics" leads to emergent phenomena of immediate significance to human beings. In real-world physics, these emergent phenomena include stars and snails; in semiotic physics, they're the stories the simulators tell and the [*simulacra*](https://www.lesswrong.com/posts/3BDqZMNSJDBg2oyvW/simulacra-are-things) that populate them. Insofar as these are unprecedented rhymes with human cognition, they merit investigation for their own sake. ([View Highlight](https://read.readwise.io/read/01k0p05102hgzvczb7ww46mp6k)) - Tags: [[ai aesthetics paper]] [[LLMs]] [[ai]] - Note: excellent statement of the idea > Taking a cue from the original [Simulators](https://www.lesswrong.com/posts/vJFdjigzmcXMhNTsx/simulators#Summary) post, which poses the question of self-supervised learning in the limit of modeling power, people sometimes ask whether the above conclusion breaks down for a sufficiently advanced simulator. At some point, this argument goes, the simulator might be able to minimize predictive loss by modeling the physical world at such a fine level of detail that humans are emulated complete with their cognitive processes. At this point, human linguistic behaviors are faithfully simulated: the simulator doesn’t need to model Harry Potter; it’s simulating the author from the physical ground up. Doesn’t this mean semiotic physics has converged to real-world physics? > The answer is no. Leaving aside the question of whether the hypothesized evolution is plausible—this [is debatable](https://www.lesswrong.com/posts/TTn6vTcZ3szBctvgb/simulators-seminar-sequence-2-semiotic-physics#fnlauiw3r0p0g)—the more important point is that even if we stipulate that it is, the conclusion still doesn’t follow, or, more precisely, doesn’t make sense. The hypothesized internalization of real-world physics would be profoundly significant, but unrelated to semiotic physics. The elementary particles and higher-level phenomena are still in disjoint universes of discourse: quarks and bosons, stars and snails (and authors) for real-world physics; tokens, stories, and simulacra for semiotic. ([View Highlight](https://read.readwise.io/read/01k0h39x5m0rw7rd1cy2e96j0a)) > It's in the nature of any analogy that the analogues are similar in some ways but not others. In this case, state changes in semiotic physics are many orders of magnitude coarser-grained (relative to the state) than those in quantum physics, the state space itself is infinitesimally smaller, the time evolution operator carries more information and more structure, and so on. We can look for hypotheses where things are similar and take caution where they're different, bearing in mind that the analogy itself is a prompt, not a theory. ([View Highlight](https://read.readwise.io/read/01k0h43afj01b38jz8drbhrewq))