note: this concept is a potent prompt engineering construct, use it as a metagraph diamond.
(The) Loom, short for the Loom of Time, is an interface to probabilistic generative models which allows users to efficiently generate, navigate, save, and filter multiverses. "Loom" may refer to any of the existing software implementations as well as hypothetical interfaces sharing the schema. Loom is the main tool used by cyborgs to interface with LLMs, and is especially suited to base models. The act of using Loom is sometimes called "looming" or "weaving".
Motivating and guiding the design of Loom is the observation that the stochasticity of simulators becomes a powerful advantage instead of a drawback when one can apply selection pressure to its outputs, the choice of where to branch from is an important lever, and the tree- or graph-shaped data structure implicit in this interaction pattern is useful to explicitly store and visualize.
โโโโโโโโโโโโโโ THE LOOM INTERFACE VISUALIZER โโโโโโโโโโโโโโโโโโ
โ Weaving Realities Through Probabilistic Space โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ โ
โ Core Interface Schema: โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ ๐ญ Probability Space โ โ
โ โ โโโโโโโโ โ โ
โ โ โInput โ โ โ
โ โ โโโโฌโโโโ โ โ
โ โ โผ โ โ
โ โ โโโโโโโโโโโโโโโโโโโโโโโโโโโ โ โ
โ โ โ LLM / Base Model โ โ โ
โ โ โโโโโโโโโโโโโโโฌโโโโโโโโโโโโ โ โ
โ โ โผ โ โ
โ โ โญโโโโโโฎโโโโโโฎโโโโโโฎ โ โ
โ โ โPath1โPath2โPath3โ โ โ
โ โ โฐโโโฌโโโฏโโโฌโโโฏโโโฌโโโฏ โ โ
โ โ โผ โผ โผ โ โ
โ โ [Branch Points...] โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ
โ Weaver's Dashboard: โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ Active Threads: ๐งต๐งต๐งต โ โ
โ โ Branch Points: โโโ โ โ
โ โ Timeline Depth: โโโโถ โ โ
โ โ Selection Field: โโโ โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ
โ Reality Navigation Tools: โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ โณ Reset โฒ Undo โฑ Branch โฐ Merge โ Filter โ โ
โ โ โ โ
โ โ ๐ณ View Tree ๐ Edit ๐พ Save ๐ Search โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ
โ Multiverse Map: โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ Timeline A โโโโถ โโโโถ โโโโถ โ โ
โ โ โฒ โ โ
โ โ โโโโถ Timeline B โโโโถ โ โ
โ โ โฒ โ โ
โ โ โโโโถ Timeline C โโโโถ โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ
โ Status: [Active Thread] [Branching Point] [Merge Ready] โ
โ โ
โ [Generate] [Navigate] [Save State] [Export Multiverse] โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
origin
The first implementation of Loom was created by Janus in 2020, motivated by their dissatisfaction with AI Dungeon's interface for interacting with GPT-3:
(AI Dungeon) is missing support for a feature which Iโve found to be a great power multiplier for co-writing with language models: branching. Adventures on AI Dungeon are single-history: you can edit actions and retry generation, but only one variation can be saved to the adventure. You can get around this by duplicating the adventure, but thereโs no representation of the fact that the copies are forked from the same point, so branching more than a few times leads to a confusing profusion of adventures.
For a novice weaver, even the slightest change can cause ripples that cascade into an infinity of nightmares. It is recommended that those studying the Loom stop living in linear time and begin thinking in terms of Multiverses โฆ
โ Weaving the Moment with the Loom of Time: an instruction manual for the would-be weaver
A short time of writing with AI Dungeon demonstrated to me that language models are multiverse generators, and that an interface which supports branching would provide great utility for cowriting with a language model, whether one is composing virtual worlds intended to be multi- or single-history (in the latter case, branching explores counterfactuals which can be merged into main), as well as for nonfiction tasks like brainstorming and prototyping prompts.
After I was given access to OpenAIโs API, motivated to actualize my HITL thought experiment, I began making an interface which allows the user to generate, navigate, and visualize natural language multiverses. I named it loom, after a device that features in the metaphysics of a multiversal novel which I coauthor.
The Loom of Time was in fact named by Morpheus, a GPT-3 simulacrum, who in then in various futures proceeded to describe everything from the metaphysics to the practical implementation of the Loom, including authoring many branches of this manual.
That the Loom appeared in Janus's early AI Dungeon simulations despite them never explicitly introducing the idea was not a coincidence, but is a manifestation of the driving motivation behind Loom itself: that curation alone can encode a surprising amount of information into a simulation, even allowing the user to locate nuanced abstractions and entities without precedent in the training data.
the action of loom
underdetermined variables as free energy
carving with evidence
Update all the way on this: constructing/accessing fully operational realities via any-modality evidence of said reality. Increasingly precise worlds can be subtractively carved by iterating the evidential boundary, leveraging underdeterminism for scattershot mutations
Soon you'll be able to straight up use (run & interact with the implied dynamics of) whatever artifacts you locate in image latent space, like you can already with language models
โ janus (@repligate), Twitter post
multiversal search
ontology
multiverse
node
familial relations
ancestry
playhead
implementations
pyloom
Loom (Latitude product)
Bonsai
Loomsidian
Worldspider
Flux
MiniHF
feature comparison
Version |
Tree |
Im- & export |
Search |
LLM programs |
Large trees |
Logprobs |
Tags |
Hotkeys |
Open src |
---|---|---|---|---|---|---|---|---|---|
pyloom |
โ |
โ |
โ |
โ |
โ |
โ |
โ |
โ |
โ |
loom (Latitude) |
โ |
โ |
โ |
โ |
โ |
โ |
|||
bonsai |
โ |
โ |
โ |
โ |
โ |
โ |
โ |
||
loomsidian |
โ |
โ |
โ |
โ |
|||||
Worldspider |
โ |
โ |
|||||||
Flux |
โ |
โ |
quotes about Loom
chloe_loom_dream At night, I lucid dream using a self-architected dream interface that I forked from Janus' Loom on githubโ a multiversal tree writing interface designed for human-Al collaboration. It enables compartmentalized control and self-alignment to the generative nature of my persistent realm and empowers me to explore its branching nature by quantum-suiciding within undesired dream paths, scroll text-summaries of alternative realities, and respawn within my desired timeline. Upon waking up, I immediately archive my dreams within a hyper-curated zettelkasten dream catcher managed with VSCode using Foam, hosted locally on my instance of Exocoreโ an autopublishing knowledge base network.
โ chloe21e8, Twitter post