Dontopedia

Micro Julia

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)

Micro Julia has 17 facts recorded in Dontopedia across 6 references.

17 facts·16 predicates·6 sources

Mostly:trained on(2), is minimal gpt in julia(1), has custom scalar autograd engine(1)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (6)

Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.

evolvesFromEvolves From(1)

evolvesFromMicroJuliaEvolves From Micro Julia(1)

isCurrentlyTrainingIs Currently Training(1)

mentionedRepositoryMentioned Repository(1)

ownsRepositoryOwns Repository(1)

ranksTopProjectsBySignificanceRanks Top Projects by Significance(1)

Other facts (17)

The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.

17 facts
PredicateValueRef
Trained onAristotles Rhetoric[4]
Trained onAristotles Rhetoric[5]
Is Minimal Gpt in Juliatrue[1]
Has Custom Scalar Autograd Enginetrue[2]
Has Parameter Count Range~5-9K params[2]
Has Single Layer Transformertrue[2]
Implemented in LanguageJulia[2]
Integrates Hugging Facetrue[2]
Integrates Wand Btrue[2]
Is Minimal Gpttrue[2]
Is Port ofKarpathy Microgpt[2]
Uses No Frameworkstrue[2]
Trained on TextAristotle's Rhetoric (cleaned from Gutenberg)[2]
Adapts Karpathy WorkKarpathy[2]
Is Unfamiliar toRolandnsharp7643[3]
Rdf:typeRepository[6]
Repository OwnerDavinci Dreams[6]

Timeline

Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.

isMinimalGptInJulialisa-watts/github-davincidreams
true
hasCustomScalarAutogradEnginelisa-watts/davincidreams-repos-deep-analysis
true
hasParameterCountRangelisa-watts/davincidreams-repos-deep-analysis
~5-9K params
hasSingleLayerTransformerlisa-watts/davincidreams-repos-deep-analysis
true
implementedInLanguagelisa-watts/davincidreams-repos-deep-analysis
ex:julia
integratesHuggingFacelisa-watts/davincidreams-repos-deep-analysis
true
integratesWandBlisa-watts/davincidreams-repos-deep-analysis
true
isMinimalGPTlisa-watts/davincidreams-repos-deep-analysis
true
isPortOflisa-watts/davincidreams-repos-deep-analysis
ex:karpathy-microgpt
usesNoFrameworkslisa-watts/davincidreams-repos-deep-analysis
true
trainedOnTextlisa-watts/davincidreams-repos-deep-analysis
Aristotle's Rhetoric (cleaned from Gutenberg)
adaptsKarpathyWorklisa-watts/davincidreams-repos-deep-analysis
ex:karpathy
isUnfamiliarToblah/vidya/part-4
ex:rolandnsharp7643
trainedOnblah/vidya/part-1
ex:aristotles-rhetoric
trainedOnblah/vidya/1
ex:aristotles-rhetoric
typeblah/vidya/5
ex:Repository
repositoryOwnerblah/vidya/5
ex:DavinciDreams

References (6)

6 references
  1. ctx:genes/lisa-watts/github-davincidreams
  2. ctx:genes/lisa-watts/davincidreams-repos-deep-analysis
  3. [3]Part 41 fact
    ctx:discord/blah/vidya/part-4
  4. [4]Part 11 fact
    ctx:discord/blah/vidya/part-1
  5. [5]11 fact
    ctx:discord/blah/vidya/1
    • full textvidya-1
      text/plain3 KBdoc:agent/vidya-1/ad39c8e8-e1cf-4553-a837-dac26303180a
      Show excerpt
      [2026-02-20 02:55] ajaxdavis: <@360636009433661440> wtf does vidya mean lol [2026-02-20 02:55] ajaxdavis: sounds like a pervert websit [2026-02-20 02:55] rolandnsharp7643: Vidya (Sanskrit: विद्या, IAST: vidyā) figures prominently in all tex
  6. [6]52 facts
    ctx:discord/blah/vidya/5
    • full textvidya-5
      text/plain2 KBdoc:agent/vidya-5/7bb294ff-00bf-4731-bfd3-215fc0850293
      Show excerpt
      [2026-02-20 11:31] rolandnsharp7643: roland@cube:~/code/flow/ocaml$ ocamlopt -O2 -o microgpt_tuned blas_stubs.c eleven_microgpt_tuned.ml -ccopt "-I/usr/include/x86_64-linux-gnu" -cclib -lopenblas && ./microgpt_tuned --load --prompt "what is

See also

Keep researching

Missing something or suspicious of what's here? Kick off a research session — a Claude agent will investigate, cite its sources, and file new facts into a dedicated context you can review before accepting into the shared view.