Dontopedia

L0

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

L0 has 22 facts recorded in Dontopedia across 6 references.

22 facts·21 predicates·6 sources

Mostly:has k(1), has r(1), precedes temporal mixing(1)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (1)

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.

hasLowerSpectralEntropyInHas Lower Spectral Entropy in(1)

Other facts (21)

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.

21 facts
PredicateValueRef
Has K1.15[1]
Has R0.23[1]
Precedes Temporal MixingTemporal Mixing[2]
Is First Layertrue[2]
Has Not Benefited FromTemporal Mixing[2]
Uses Different Row Offsetsnull[3]
PlacesParents Weights[3]
SharesInput Dim[3]
Amplifies Input by860%[4]
Has Mean Input Norm2.65[4]
Has Mean Block Out Norm22.76[4]
Handles Encoding BlowupHaar Encoding[4]
Has Ratio Percent860[4]
AmplifiesHaar Encoding[4]
Has K Value1.15[5]
Has R Value0.23[5]
Rdf:typeModel Layer[6]
Contribution Ratio1.57[6]
Layer Roleinput-conditioner[6]
Dominates LayersLayers 1 to 19[6]
Adjacency Sparsity0.04[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.

hasKblah/watt-activation/part-8
1.15
hasRblah/watt-activation/part-8
0.23
precedesTemporalMixingblah/watt-activation/part-452
ex:temporal-mixing
isFirstLayerblah/watt-activation/part-452
true
hasNotBenefitedFromblah/watt-activation/part-452
ex:temporal-mixing
usesDifferentRowOffsetsblah/watt-activation/part-490
null
placesblah/watt-activation/part-490
ex:parents-weights
sharesblah/watt-activation/part-490
ex:input-dim
amplifiesInputByblah/watt-activation/part-684
860%
hasMeanInputNormblah/watt-activation/part-684
2.65
hasMeanBlockOutNormblah/watt-activation/part-684
22.76
handlesEncodingBlowupblah/watt-activation/part-684
ex:haar-encoding
hasRatioPercentblah/watt-activation/part-684
860
amplifiesblah/watt-activation/part-684
ex:haar-encoding
hasKValueblah/watt-activation/8
1.15
hasRValueblah/watt-activation/8
0.23
typeblah/watt-activation/683
ex:ModelLayer
labelblah/watt-activation/683
L0
contributionRatioblah/watt-activation/683
1.57
layerRoleblah/watt-activation/683
input-conditioner
dominatesLayersblah/watt-activation/683
ex:layers-1-to-19
adjacencySparsityblah/watt-activation/683
0.04

References (6)

6 references
  1. [1]Part 82 facts
    ctx:discord/blah/watt-activation/part-8
  2. [2]Part 4523 facts
    ctx:discord/blah/watt-activation/part-452
  3. [3]Part 4903 facts
    ctx:discord/blah/watt-activation/part-490
  4. [4]Part 6846 facts
    ctx:discord/blah/watt-activation/part-684
  5. [5]82 facts
    ctx:discord/blah/watt-activation/8
    • full textwatt-activation-8
      text/plain3 KBdoc:agent/watt-activation-8/c065f0d3-ec70-4dc8-b305-5fd7a4e6d77a
      Show excerpt
      [2026-02-27 01:08] xenonfun: ``` ### implement phase-gated residual (#5): x + r * h instead of x + h in KuramotoBlock This is great data. At epoch 9 it's already at PPL 86.11 — matching the no-FFN baseline's final PPL of 86.09, but: - n
  6. [6]6836 facts
    ctx:discord/blah/watt-activation/683
    • full textwatt-activation-683
      text/plain3 KBdoc:agent/watt-activation-683/1d89c3e1-d173-4432-968b-898b740f9ed3
      Show excerpt
      [2026-04-23 17:37] xenonfun: All 20 layers healthy — no issues. - Zero dead layers. Contribution ratio range: 34-157% (dead threshold is <1%). L0 dominates (157%) as expected input-conditioner; L1-L19 all 34-94%. - No gate collapse. α

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.