Ask
the work.

CONCEPT / DISCOVERY ASSOCIATIONS

11 records tagged model as database. These are discovery associations from editorial tags or film metadata, not evidence that the records agree.

AUTOMATIC CAPTIONS / 0:00

What if I said every single large language model was a database? Specifically, a graph database. Not metaphorically, but physically. If we looked underneath the hood at the weights, it was an actual database with real edges, with real entities, with real nodes. That's how it was physically represented. What would that mean? Well, it would mean that we could query it. We

AUTOMATIC CAPTIONS / 0:25

could insert new knowledge. We could compile down to weights because it's just a database. And if it's just a database, then we could have query languages like SQL over the top of it. We could programmatically work with the model if that were true. But the good news is it is true. So, let's do it. So, we're going to query it with a language called Larql, which allows us to query

EDITORIAL DRAFT / OBSERVATION · OPEN · V0.2

05 / THE MODEL HAS TO BUILD THE QUERY. The film asks for Australia’s capital. Sydney becomes prominent around layer 24; Canberra leads by layer 26 and the answer settles later. These are intermediate readouts during one prediction, rather than words the model has already printed. The state reaching each FFN changes as attention and earlier FFNs update it. Addressing is therefore a process across layers. This example motivates inspecting when a useful query becomes available; the picture alone does not prove an exact database lookup.

EDITORIAL DRAFT / OBSERVATION · ONGOING · V0.1

THE THREAD. A model-as-database proposition connects this work to VINDEX3. The context notebook asks what becomes possible when earlier state can be returned to, rather than repeatedly reconstructed.

EDITORIAL DRAFT / STATEMENT · ONGOING · V0.1

The model is the database.

EDITORIAL DRAFT / COMPARISON · ONGOING · V0.1

A model — a conceptual comparison. As a file: An object to load; Addressed from the outside. As a database: Parts to address; Relationships to inspect.

EDITORIAL DRAFT / OBSERVATION · OPEN · V0.2

01 / WHAT IS AN ADDRESS?. Ask for the capital of France. France identifies the entity; capital identifies the relation you want to read. Currency would ask something different about the same place. In the opening animation, that question arrives at an FFN and a Paris direction is written into the residual stream—the model’s changing working state. The conveyor is an explanatory drawing. The address in a model is a vector of numbers, and matching it need not isolate a single neuron or a single fact.

EDITORIAL DRAFT / OBSERVATION · OPEN · V0.1

01 / TWO CLAIMS IN ONE CONTEXT. Any memory system that summarises, supersedes or evicts eventually holds two things that disagree: an original passage and a shorter record standing in for it. The engineering question sounds like bookkeeping — which one is authoritative? The measured answer is not a preference the model expresses. It is a property of what the model can still read at the moment it is asked.

EDITORIAL DRAFT / EVIDENCE · OPEN · V0.1

PROMOTION / THREE OPERATIONS [SUPPORTED]. In the founding run, a 13-token record injected about 49,000 tokens downstream of the source keeps the model correct on all three tested operations — copy 7431, plus one 7432, digit reversal 1347 — with the original span excluded. Padding alone fails on every row (7249, 789, 7892) and a corrupted record steers every row (5824, 5825, 4285). Asked for the code plus one with 5824 injected, the model answers 5825: a value that appears nowhere in the context.

EDITORIAL DRAFT / STATEMENT · ONGOING · V0.1

Models are places you can go.

EDITORIAL DRAFT / STATEMENT · OPEN · V0.1

Context is treated as expensive because we repeatedly reconstruct it.

EDITORIAL DRAFT / OBSERVATION · OPEN · V0.1

Every new call re-derives what a previous call already established, at real token cost, because nothing kept the earlier state in a form worth returning to. If that state were indexed rather than discarded, context stops being the scarce resource everyone budgets around.

TITLE MATCH / TRANSCRIPT NOT INDEXED · V1.0

superduperdb supercharges your database for AI

TITLE MATCH / TRANSCRIPT NOT INDEXED · V1.0

Getting started with SurrealDB!! Future of cloud databases (maybe)?

TITLE MATCH / TRANSCRIPT NOT INDEXED · V1.0

intro to graph databases

TITLE MATCH / TRANSCRIPT NOT INDEXED · V1.0

an introduction to neo4j (graph database tutorial for beginners)

CURATED THREAD / WORKING EDITION · V0.1

What can a model’s changing state tell us about the way it reads? Follow the map film through the visual notebooks, studies you can work yourself, and the questions behind LARQL and VINDEX3. A curated reading order, rather than a chronology of discovery. Films keep their original dates. Notes and dossiers remain working drafts. Interactive studies are labelled as either constructed examples or replays of recorded results. WATCH THE STATE MOVE. Begin with the capital of Japan. The film follows the changing residual state, then asks what happens when a France state enters an Australia computation. ASK WHAT THE PICTURE MEANS. Which token position is moving? What are the axes? The notebook reads the demonstration alongside its projection code and distinguishes a useful picture from an answer probability. CHANGE THE STATE. Replacing one position and replacing the full sequence are different interventions. Follow the transplant evidence into the question of what must persist after the next word. EXAMINE THE READER. A state becomes useful when an operation can read it. Follow key matches, activations and value contributions, then compare the constructed memory with the native-model experiments. MAKE AN INTERVENTION. Choose capital of Atlantis. Inspect the competing activations, remove a neuron and watch the answer scores change. Every operation in this little memory is visible. ASK WHICH ONE ANSWERS. A stored memory eventually holds two things that disagree. Follow the recorded arc from a record that replaces a retired source, to one that is completely inert against a source still being read. RETIRE A READ. Eight attention layers can see the whole context. Switch one off and the newer record takes the answer. The study replays what was measured, and stays silent about what was not. WATCH THE ADDRESS TAKE SHAPE. Return to native facts. A final-position transplant and an independent reader locate the entity transition between L24 and L28. Explore the early relation frame and changing coordinates, with the answer-token endpoint marked explicitly. TURN THE QUESTION TOWARDS SOFTWARE. LARQL explores how learned systems can be queried. Return from the small memory to the larger engineering question: which operations can a model reliably expose? ASK HOW TO REPRESENT IT. If a model has parts we can address and relationships we can inspect, how should we represent them? VINDEX3 gives that question its own system and specification.

TITLE & DESCRIPTION · V1.0

Quantization is usually described as if a model simply goes from 16-bit to 8-bit to 4-bit. But real model representations are more complicated than that: different tensors can use different formats, scales, group sizes, packing schemes and precision choices — and two things described as “4-bit” may behave very differently.