Dontopedia

small

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-10.)

small has 18 facts recorded in Dontopedia across 8 references, with 2 live disagreements.

18 facts·15 predicates·8 sources·2 in dispute

Mostly:rdf:type(2), has speedup factor(2), is trained from ground up(1)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (8)

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.

aimsToReplicateWithAims to Replicate With(1)

categorizesAsNotWorkingYetCategorizes As Not Working Yet(1)

hasModelVariantHas Model Variant(1)

improvesBPBOfImproves Bpb of(1)

notablyImprovesPerformanceNotably Improves Performance(1)

rdf:typeRdf:type(1)

variantVariant(1)

worksCorrectlyWorks Correctly(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
Rdf:typeModel Variant[5]
Rdf:typeModel Size Category[8]
Has Speedup Factor1.2[7]
Has Speedup Factor1.6[7]
Is Trained From Ground UpFp4[1]
Is Designed forFp4[1]
Exhibits Speedup Growth1.2x → 1.6x[2]
Expected BehaviorSpeedup Growth[2]
Cannot FormFine Grained Distinctions[3]
Has DimensionD 128[3]
Lacks Capacity forFine Grained Distinctions[3]
Refers toGated Rotor Bpsk[4]
Checkpoint Size9.7MB[6]
Parameter M31[6]
Layer Size L66[6]
Has Speedup CharacteristicGrowth With Length[7]
Speedup Measured at Token Count120[7]

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.

18 undated
—
isTrainedFromGroundUpblah/general/part-129
ex:fp4
—
isDesignedForblah/general/part-129
ex:fp4
—
exhibitsSpeedupGrowthblah/watt-activation/part-99
1.2x → 1.6x
—
expectedBehaviorblah/watt-activation/part-99
ex:speedup-growth
—
cannotFormblah/watt-activation/part-286
ex:fine-grained-distinctions
—
hasDimensionblah/watt-activation/part-286
ex:d-128
—
lacksCapacityForblah/watt-activation/part-286
ex:fine-grained-distinctions
—
refersToblah/watt-activation/part-404
ex:gated-rotor-bpsk
—
typebeam/a872d40f-b4d5-484d-a8f6-549d5ae5c2f8
ex:ModelVariant
—
labelbeam/a872d40f-b4d5-484d-a8f6-549d5ae5c2f8
small
—
checkpointSizeblah/random/25
9.7MB
—
parameterMblah/random/25
31
—
layerSizeLblah/random/25
66
—
hasSpeedupCharacteristicblah/watt-activation/99
ex:growth-with-length
—
hasSpeedupFactorblah/watt-activation/99
1.2
—
hasSpeedupFactorblah/watt-activation/99
1.6
—
speedupMeasuredAtTokenCountblah/watt-activation/99
120
—
typebeam/d54c1b34-b976-4b4c-9900-18fb5cd506dc
ex:ModelSizeCategory

References (8)

8 references
  1. [1]Part 1292 facts
    ctx:discord/blah/general/part-129
  2. [2]Part 992 facts
    ctx:discord/blah/watt-activation/part-99
  3. [3]Part 2863 facts
    ctx:discord/blah/watt-activation/part-286
  4. [4]Part 4041 fact
    ctx:discord/blah/watt-activation/part-404
  5. [5]A872d40f B4d5 484d A8f6 549d5ae5c2f82 facts
    ctx:claims/beam/a872d40f-b4d5-484d-a8f6-549d5ae5c2f8
    • full textbeam-chunk
      text/plain1 KBdoc:beam/a872d40f-b4d5-484d-a8f6-549d5ae5c2f8
      Show excerpt
      This function and calculation should accurately reflect the cost based on the provided parameters. If you have any additional requirements or need further adjustments, feel free to ask! [Turn 2442] User: I want to research options for my t…
  6. [6]253 facts
    ctx:discord/blah/random/25
    • full textrandom-25
      text/plain2 KBdoc:agent/random-25/d9947cdf-8a50-4fe9-9125-a458363f2d30
      Show excerpt
      [2026-02-17 09:28] xenonfun: total, the cpu side needs about 600. and the bpe process is on CPU tho it got it multithreaded so its about 15s then saves vocab. it should ahve 1K run done in 30ish minutes to apples to apples compare, should b…
  7. [7]994 facts
    ctx:discord/blah/watt-activation/99
    • full textwatt-activation-99
      text/plain3 KBdoc:agent/watt-activation-99/2710131f-bb93-45bc-82d2-b9c6a42b0fea
      Show excerpt
      [2026-03-08 05:53] ajaxdavis: you are going to post train the chatty on yeah [2026-03-08 05:55] xenonfun: yeah I would try fine tuning that in or renforcement learn it (I get all the lora/dora and think we also had renforcement learning fro…
  8. [8]D54c1b34 B976 4b4c 9900 18fb5cd506dc1 fact
    ctx:claims/beam/d54c1b34-b976-4b4c-9900-18fb5cd506dc
    • full textbeam-chunk
      text/plain1 KBdoc:beam/d54c1b34-b976-4b4c-9900-18fb5cd506dc
      Show excerpt
      [Turn 9874] User: I'm designing a modular flow for query rewriting to process 2,000 queries/sec with 99.8% uptime, and I want to use spaCy 3.7.2 for tokenization, but I'm not sure how to integrate it with my existing pipeline - can you prov…

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.

↑