CHRIS HAY / FOLLOW THE THREAD
From a map
to a memory.
What can a model’s changing state tell us about the way it reads? Follow the map film into three visual notebooks, a memory you can change 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; the interactive study is a constructed example.
- 01film
WATCH THE STATE MOVE
370,000 tokens loaded in Context in 2.8MB, on a MacBook.
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.
ORIGINAL FILM
WATCH FROM 02:00 ↗
FROM THE FILM ↗ - 02notebook
ASK WHAT THE PICTURE MEANS
What is the map actually a map of?
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.
WORKING DRAFT · OPEN
OPEN THE NOTE ↗
CONSTRUCTED VISUAL STUDY ↗ - 03notebook
CHANGE THE STATE
What has to survive?
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.
WORKING DRAFT · OPEN
OPEN THE NOTE ↗
FROM THE FILM ↗ - 04notebook
EXAMINE THE READER
Reading by address.
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.
WORKING DRAFT · OPEN
OPEN THE NOTE ↗
CONSTRUCTED VISUAL STUDY ↗ - 05interactive study
MAKE AN INTERVENTION
Try the mechanism.
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.
CONSTRUCTED EXAMPLE
TRY THE MECHANISM ↗CAPITAL OF ATLANTISParis.CHANGE THE QUESTION ↗ - 06work
TURN THE QUESTION TOWARDS SOFTWARE
LARQL
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?
WORKING DRAFT · ONGOING
EXPLORE THE WORK ↗
FROM THE FILM ↗ - 07workCONCEPTUAL STUDY ↗
ASK HOW TO REPRESENT IT
VINDEX3
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.
WORKING DRAFT · ONGOING
EXPLORE THE WORK ↗
THE NEXT QUESTION
What would make
the address reliable?
The notes keep the limits visible: changed phrasing, competing entities, retained state and use beyond the first answer. Follow the open research question, or search the underlying records.