manifold
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
manifold has 13 facts recorded in Dontopedia across 9 references, with 2 live disagreements.
Mostly:rdf:type(3), underlies dynamics(1), native for(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (22)
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.
staysOnStays on(3)
- Computation
ex:computation - Low Rank Cross Attention
ex:low-rank-cross-attention - Manifold Preserving Low Rank Cross Attention on S H 1
ex:manifold-preserving-low-rank-cross-attention-on-s-h-1
notDerivedFromNot Derived From(2)
- L1 Normalization
ex:l1-normalization - L1 Normalization
ex:l1-normalization
attributedLearningToAttributed Learning to(1)
- Lisamegawatts
ex:lisamegawatts
carriedPassengerCarried Passenger(1)
- Quiraing
ex:quiraing
enablesMoreEfficientSearchEnables More Efficient Search(1)
- Manifold Unit
ex:manifold-unit
freesEmbedWeightFrees Embed Weight(1)
- Option 3 Free Weights
ex:option-3-free-weights
freesLmHeadWeightFrees Lm Head Weight(1)
- Option 3 Free Weights
ex:option-3-free-weights
illegitimatePurposesIllegitimate Purposes(1)
- Bathing House
ex:bathing-house
includesIncludes(1)
- Underlying Geometric Structure
ex:underlying-geometric-structure
isIs(1)
- Ofdm Wire Encoding
ex:ofdm-wire-encoding
isProvablyOnIs Provably on(1)
- Cayley U N Rotor Dynamics
ex:cayley-u-n-rotor-dynamics
keptNearKept Near(1)
- System
ex:system
lacksConstraintLacks Constraint(1)
- Adam
ex:adam
locatedAtLocated at(1)
- Mining Manager Reid
ex:mining-manager-reid
locatedOnLocated on(1)
- Finite Attractor Mechanism
ex:finite-attractor-mechanism
locationLocation(1)
- Mining Manager Kidd
ex:mining-manager-kidd
preservesPreserves(1)
- Low Rank Cross Attention
ex:low-rank-cross-attention
shouldStayOnShould Stay on(1)
- Computation
ex:computation
workedOnManuallyWorked on Manually(1)
- Xenonfun Session 1
ex:xenonfun-session-1
Other facts (9)
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.
| Predicate | Value | Ref |
|---|---|---|
| Rdf:type | Mathematical Object | [6] |
| Rdf:type | Mathematical Structure | [7] |
| Rdf:type | Model Component | [9] |
| Underlies Dynamics | Lohe Dynamics | [1] |
| Native for | Phases | [2] |
| Is Correct for DC Neutral Filters | true | [3] |
| Is Repeatedly Emphasized | Get Back on Manifold | [4] |
| Has Params | 0.027M | [5] |
| Has Per Layer Groups | 16 | [5] |
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.
References (9)
ctx:discord/blah/watt-activation/part-137ctx:discord/blah/watt-activation/part-321ctx:discord/blah/watt-activation/part-352ctx:discord/blah/watt-activation/part-692ctx:discord/blah/watt-activation/part-699ctx:discord/blah/watt-activation/137- full textwatt-activation-137text/plain2 KB
doc:agent/watt-activation-137/9608fbdb-8ce7-4e5b-ac20-5c329d46eadeShow excerpt
[2026-03-09 06:07] xenonfun: ``` ❯ are you sure this is principaled? ⏺ No, you're right to push back. L1 normalization is ad-hoc — it bounds the magnitude but it's not derived from the manifold. The principled fix is to normalize the mean…
ctx:discord/blah/watt-activation/343- full textwatt-activation-343text/plain2 KB
doc:agent/watt-activation-343/758b6afd-7936-40ff-806a-3684984cdf9eShow excerpt
[2026-03-15 23:13] xenonfun: ``` ⏺ Pushed f5d17e5. All EMA/stability/spike hacks removed. Both adaptive LR modes are now pure physics: direct measurement × capacity theory × convergence decay. On the LoheCrossCoupleModes comment — that's…
ctx:discord/blah/watt-activation/345- full textwatt-activation-345text/plain3 KB
doc:agent/watt-activation-345/c59946eb-7ad9-465b-939c-f70436033800Show excerpt
[2026-03-16 01:39] xenonfun: ⏺ Yes — principled noise injection is exactly what communications systems do. Three reasons it could help: 1. Stochastic resonance. In nonlinear systems (which Lohe sync IS), a small amount of noise can actua…
ctx:discord/blah/watt-activation/620- full textwatt-activation-620text/plain2 KB
doc:agent/watt-activation-620/76566bac-aa9d-4345-b1ab-afded7e71af5Show excerpt
[2026-04-11 10:36] lisamegawatts: The loss dropped from ~36 to 9.3 in 100 steps — the manifold is learning! The output went from random multilingual tokens to English words with approximate semantic relevance ("technology", "processing", "e…
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.