Dontopedia

4-layer (5.1M)

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

4-layer (5.1M) has 11 facts recorded in Dontopedia across 1 reference.

11 facts·10 predicates·1 sources

Mostly:layer count(1), parameter count(1), rdf:type(1)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (3)

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.

comparedToCompared to(1)

contrastsStateContrasts State(1)

involvesComparisonInvolves Comparison(1)

Other facts (10)

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.

10 facts
PredicateValueRef
Layer Count4[1]
Parameter Count5100000[1]
Rdf:typeModel Configuration[1]
Final Avg Perplexity72.2[1]
Best Perplexity9[1]
Throughput117[1]
Final Correlation0.85[1]
Training Time171[1]
Learning StatusConverging[1]
Has InferenceXenonfun[1]

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.

labelblah/watt-activation/50
4-layer (5.1M)
layerCountblah/watt-activation/50
4
parameterCountblah/watt-activation/50
5100000
typeblah/watt-activation/50
ex:ModelConfiguration
finalAvgPerplexityblah/watt-activation/50
72.2
bestPerplexityblah/watt-activation/50
9
throughputblah/watt-activation/50
117
finalCorrelationblah/watt-activation/50
0.85
trainingTimeblah/watt-activation/50
171
learningStatusblah/watt-activation/50
ex:converging
hasInferenceblah/watt-activation/50
ex:xenonfun

References (1)

1 references
  1. [1]5011 facts
    ctx:discord/blah/watt-activation/50
    • full textwatt-activation-50
      text/plain2 KBdoc:agent/watt-activation-50/f0e77349-2ac7-422c-8f3c-c44c6fba499b
      Show excerpt
      [2026-03-07 07:00] xenonfun: ``` 8-layer results: ┌───────────────┬────────────────┬─────────────────┐ │ Metric │ 4-layer (5.1M) │ 8-layer (11.5M) │ ├───────────────┼────────────────┼─────────────────┤ │ Final avg PPL │ 72.2

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.