Dontopedia

lohe_complex_split

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

lohe_complex_split has 21 facts recorded in Dontopedia across 6 references, with 4 live disagreements.

21 facts·15 predicates·6 sources·4 in dispute

Mostly:has learned(2), applies to(2), has parameter(2)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (12)

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.

consistsOfConsists of(2)

appliesToApplies to(1)

discussesTopicDiscusses Topic(1)

implementsVariantImplements Variant(1)

includesComponentIncludes Component(1)

isMoreStructuredThanIs More Structured Than(1)

isSuperiorToIs Superior to(1)

makesContentAdaptiveMakes Content Adaptive(1)

referencesReferences(1)

usesArchitectureFeatureUses Architecture Feature(1)

usesAttentionTypeUses Attention Type(1)

Other facts (19)

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.

19 facts
PredicateValueRef
Has LearnedBase Freq[2]
Has LearnedDispersion Slope[2]
Applies toComplex Gamma[2]
Applies toAttention Mechanism[3]
Has ParameterBase Freq[4]
Has ParameterDispersion Slope[4]
Rdf:typeAttention Mechanism[5]
Rdf:typeFeature[6]
Learns Frequencies Independentlytrue[1]
Applies Same Filter to Every PositionSequence Position[1]
Presupposes Existence inAudience System[1]
Has Learned Constant Poletrue[1]
Adds Phase AngleBandpass Filter[1]
Overlaps Partially WithHelmholtz Dynamics[1]
Has Per Head Complex Polestrue[1]
Has FixedMagnitude Decay Per Head[2]
Has FeatureComplex Poles With Dispersion[3]
Decay CharacteristicMagnitude Decay Fixed Per Head[4]
Statusvalidated[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.

21 undated
—
learnsFrequenciesIndependentlyblah/random/part-39
true
—
appliesSameFilterToEveryPositionblah/random/part-39
ex:sequence-position
—
presupposesExistenceInblah/random/part-39
ex:audience-system
—
hasLearnedConstantPoleblah/random/part-39
true
—
addsPhaseAngleblah/random/part-39
ex:bandpass-filter
—
overlapsPartiallyWithblah/random/part-39
ex:helmholtz-dynamics
—
hasPerHeadComplexPolesblah/random/part-39
true
—
hasLearnedblah/watt-activation/part-405
ex:base-freq
—
appliesToblah/watt-activation/part-405
ex:complex-gamma
—
hasLearnedblah/watt-activation/part-405
ex:dispersion-slope
—
hasFixedblah/watt-activation/part-405
ex:magnitude-decay-per-head
—
hasFeatureblah/random/40
ex:complex-poles-with-dispersion
—
appliesToblah/random/40
ex:attention-mechanism
—
hasParameterblah/watt-activation/403
ex:base-freq
—
hasParameterblah/watt-activation/403
ex:dispersion-slope
—
decayCharacteristicblah/watt-activation/403
ex:magnitude-decay-fixed-per-head
—
typeblah/watt-activation/423
ex:AttentionMechanism
—
labelblah/watt-activation/423
lohe_complex_split
—
typeblah/watt-activation/475
ex:Feature
—
labelblah/watt-activation/475
lohe_complex_split
—
statusblah/watt-activation/475
validated

References (6)

6 references
  1. [1]Part 397 facts
    ctx:discord/blah/random/part-39
  2. [2]Part 4054 facts
    ctx:discord/blah/watt-activation/part-405
  3. [3]402 facts
    ctx:discord/blah/random/40
    • full textrandom-40
      text/plain3 KBdoc:agent/random-40/730d8c77-858f-41f8-a687-36cfdb4d5cc9
      Show excerpt
      [2026-03-19 00:22] xenonfun: ``` The open question: Your existing lohe_complex_split attention already has complex poles with dispersion. The Helmholtz fiber is doing something very similar but at the FFN level rather than the attention…
  4. [4]4033 facts
    ctx:discord/blah/watt-activation/403
    • full textwatt-activation-403
      text/plain2 KBdoc:agent/watt-activation-403/d55f91e3-d6ae-47d0-a701-fcf659140c12
      Show excerpt
      [2026-03-19 05:31] xenonfun: ❯ is any of this applicable to WireLM lohe_v3 style stuff? ``` ⏺ Yes — the decay gates specifically. The gated cumsum monoid (g, v) ⊗ (g', v') = (g·g', g'·v + v') is exactly what _gated_cumsum already does in at…
  5. [5]4232 facts
    ctx:discord/blah/watt-activation/423
    • full textwatt-activation-423
      text/plain2 KBdoc:agent/watt-activation-423/8045fe79-be14-40ef-8627-1f24b38b2ba2
      Show excerpt
      [2026-03-20 01:28] xenonfun: ``` ⏺ OK, so for H≠4 it dispatches to _spectral_lohe_sync in attention.py (line 162), and for H=4 it uses quaternion_lohe_sync. Our entity run uses H=2, so it wouldn't even hit the Lohe FFN path — ResonantWir…
  6. [6]4753 facts
    ctx:discord/blah/watt-activation/475
    • full textwatt-activation-475
      text/plain2 KBdoc:agent/watt-activation-475/e1172583-2318-4515-b9d4-b2208603f689
      Show excerpt
      [2026-03-21 20:42] xenonfun: ``` ⏺ Clean tree, all pushed. Here's the full session summary: Session Summary: 15 commits, ~20K new lines FedSym Evolutionary Framework (complete): - 15 modules: config, activation, unit, training, fusi…

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.

↑