Dontopedia

GPT-2 Scale

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

GPT-2 Scale has 3 facts recorded in Dontopedia across 3 references.

3 facts·2 predicates·3 sources
Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (14)

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.

atScaleAt Scale(1)

dominatesComputeDominates Compute(1)

empiricallyFailsEmpirically Fails(1)

evaluatedAtEvaluated at(1)

evaluatedNegativelyEvaluated Negatively(1)

evaluatesLocalTrainingAsAbsurdlyCheapEvaluates Local Training As Absurdly Cheap(1)

expectsSmallerThroughputGapExpects Smaller Throughput Gap(1)

failsAtScaleFails at Scale(1)

isOnlyOptimizerLearningIs Only Optimizer Learning(1)

mentionsEntityMentions Entity(1)

notUsefulAtNot Useful at(1)

operatesAtScaleOperates at Scale(1)

preferredOverOthersPreferred Over Others(1)

referencesGpt2References Gpt2(1)

Other facts (2)

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.

2 facts
PredicateValueRef
Presupposes Data SizeFineweb Edu[1]
Contextually Large Model Size37m Parameters[2]

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.

3 undated
—
presupposesDataSizeblah/watt-activation/part-119
ex:fineweb-edu
—
contextuallyLargeModelSizeblah/watt-activation/part-183
ex:37m-parameters
—
labelblah/watt-activation/83
GPT-2 Scale

References (3)

3 references
  1. [1]Part 1191 fact
    ctx:discord/blah/watt-activation/part-119
  2. [2]Part 1831 fact
    ctx:discord/blah/watt-activation/part-183
  3. [3]831 fact
    ctx:discord/blah/watt-activation/83
    • full textwatt-activation-83
      text/plain2 KBdoc:agent/watt-activation-83/d1951553-a3a8-485d-8f52-ce6fc526092a
      Show excerpt
      [2026-03-07 19:38] xenonfun: ``` 128K context — clean pass. ┌─────────┬───────┬───────┬────────┬────────┬─────────┐ │ Seq Len │ it/s │ tok/s │ Active │ Peak │ Compile │ ├─────────┼───────┼───────┼────────┼────────┼─────────┤ │ L…

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.

↑