oscillator dynamics
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
oscillator dynamics has 44 facts recorded in Dontopedia across 17 references, with 4 live disagreements.
Mostly:includes(6), rdf:type(4), responds to(3)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (18)
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.
basedOnBased on(2)
- Dynamic Step Size Setting
ex:dynamic-step-size-setting - Manifold Unit S³
ex:ManifoldUnit-S³
dependsDynamicallyOnDepends Dynamically on(2)
- Ffn T Step Size
ex:ffn-t-step-size - R
ex:r
shouldControlShould Control(2)
- R Adaptive Lr
ex:r-adaptive-lr - R Adaptive Lr
ex:r-adaptive-lr
comparesToCompares to(1)
- Kan Attention
ex:kan-attention
decaysDecays(1)
- Vector
ex:vector
describesStateDescribes State(1)
- Message 2
ex:message-2
enablesRicherOscillatorDynamicsEnables Richer Oscillator Dynamics(1)
- Model
ex:model
growsGrows(1)
- Bivector
ex:bivector
isDynamicBasedOnIs Dynamic Based on(1)
- R
ex:r
isSimplerThanIs Simpler Than(1)
- Kan Attention
ex:kan-attention
isSuperiorInIs Superior in(1)
- Dyn H4 Phase Run
ex:dyn-h4-phase-run
removingFixedCouplingDramaticallyImprovesRemoving Fixed Coupling Dramatically Improves(1)
- Dyn H4 Phase Run
ex:dyn-h4-phase-run
showsConsistentlyBetterShows Consistently Better(1)
- Phase Coupling
ex:phase-coupling
superiorInSuperior in(1)
- Phase Coupling
ex:phase-coupling
usesUses(1)
- Byte Level Models
ex:byte-level-models
Other facts (40)
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 |
|---|---|---|
| Includes | Manifold Projection | [7] |
| Includes | Coupling | [7] |
| Includes | Sync | [7] |
| Includes | Sync | [14] |
| Includes | Coupling | [14] |
| Includes | Manifold Projection | [14] |
| Rdf:type | Concept | [10] |
| Rdf:type | Dynamics | [12] |
| Rdf:type | Mechanism | [15] |
| Rdf:type | Computational Method | [16] |
| Responds to | Channel Noise Bandwidth | [3] |
| Responds to | Channel Noise | [12] |
| Responds to | Bandwidth | [12] |
| Has Aspect | coupling | [13] |
| Has Aspect | decay | [13] |
| Add Overhead | null | [1] |
| Are Overhead | null | [1] |
| Unnecessary for End Product | Kan Attention | [2] |
| Responding to | Channel Noise Bandwidth | [3] |
| Includes Coupling and Decay | null | [4] |
| Structured by | Phase Dependent Coupling | [5] |
| Improved Dramatically in | Dyn H4 Phase Run | [6] |
| Key Performance Aspect | true | [6] |
| Classified As | Activation Functions | [7] |
| Not Fuseable As | Weights | [7] |
| Occur Between | Linear Layers | [7] |
| Distinguishes Which Fragment Comes Next | True | [8] |
| Provide Enough Context Dependent State | True | [8] |
| Are Visible | Simulation | [9] |
| Are Working Correctly | null | [9] |
| Location Is | Between Linear Layers | [14] |
| Functions As | Activation Function | [14] |
| Not Fusable | true | [14] |
| Provides | context-dependent state | [15] |
| Occurs on | Sphere | [16] |
| Has Scalar Behavior | Scalar Oscillation 7.5 to 7.3 | [17] |
| Has Vector Behavior | Vector Decay | [17] |
| Has Bivector Behavior | Bivector Growth | [17] |
| Has Trivector Value | 0 | [17] |
| Has Cause | Commutator Limitation | [17] |
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 (17)
ctx:discord/blah/watt-activation/part-66ctx:discord/blah/watt-activation/part-63ctx:discord/blah/watt-activation/part-349ctx:discord/blah/watt-activation/part-393ctx:discord/blah/watt-activation/part-426ctx:discord/blah/watt-activation/part-428ctx:discord/blah/watt-activation/part-437ctx:discord/blah/watt-activation/part-470ctx:discord/blah/watt-activation/part-503ctx:discord/blah/watt-activation/63- full textwatt-activation-63text/plain2 KB
doc:agent/watt-activation-63/1bd53136-248b-4353-b53c-b8c81b9d26f4Show excerpt
[2026-03-07 15:30] xenonfun: ### no GPU contention run ``` Clean results at 5K iters: ┌───────────────────┬──────┬───────┬──────────┬───────────┬──────────┬───────┬─────────┐ │ Config │ it/s │ tok/s │ Avg Loss │ Final PPL …
ctx:discord/blah/watt-activation/337- full textwatt-activation-337text/plain2 KB
doc:agent/watt-activation-337/5411616d-5166-4f26-9baf-0251a9dfbd3eShow excerpt
[2026-03-15 16:45] xenonfun: Inference: **The quick brown foxes** as the Bark Fades Are where there are also read to make a fortage of all the region for training system and made up for making the dream and object with the three kind …
ctx:discord/blah/watt-activation/347- full textwatt-activation-347text/plain2 KB
doc:agent/watt-activation-347/f5c35cb2-4a8a-467c-ac77-0754bba63d48Show excerpt
[2026-03-16 15:40] xenonfun: trying to get dynamics we wanted from start to really work, fact is its been doing very well and most of signal has been getting smashed. finally free of normalization in layers, that is where our amplitude sign…
ctx:discord/blah/watt-activation/391- full textwatt-activation-391text/plain2 KB
doc:agent/watt-activation-391/66b2bebc-08da-4969-bcc4-21de184bbdbdShow excerpt
[2026-03-19 03:37] xenonfun: ``` ⏺ Got the rotor results from the earlier run (before the r-adaptive LR change): ┌───────────────────┬──────────┬───────┬───────┬────────┐ │ Model │ BPB @500 │ tok/s │ r │ Params │ ├───…
ctx:discord/blah/watt-activation/435- full textwatt-activation-435text/plain2 KB
doc:agent/watt-activation-435/6e80af4f-2aed-449f-9f6a-4750597bfb8eShow excerpt
[2026-03-20 06:59] xenonfun: ``` ⏺ You're right. The PR #7 results (p=0.005, d=2.57, 5/5 seeds) were validated with a specific fusion operator — _block_diagonal_transfer() on vanilla nn.Linear layers. The fusion level (how weights map be…
ctx:discord/blah/watt-activation/468- full textwatt-activation-468text/plain3 KB
doc:agent/watt-activation-468/164171db-4232-40af-8c94-36b0b47c762aShow excerpt
[2026-03-21 18:49] xenonfun: ``` ├───────────────────────┼─────────────────────────┼──────────────────────────────────┼───────────────────┤ │ 3. Multi-seed MNIST │ 0.930 ± 0.019 train │ 0.995 ± 0.005 train │ CONFIRMED …
ctx:discord/blah/watt-activation/478- full textwatt-activation-478text/plain2 KB
doc:agent/watt-activation-478/0051ed8f-09c6-4572-934e-351b3f5ac139Show excerpt
[2026-03-21 22:14] xenonfun: ``` ⏺ 96.3% accuracy on MNIST with ManifoldUnit S³ — 100 iterations, 3 minutes. That's essentially matching Python's CNN result (96.2%) with no convolutions at all, just oscillator dynamics on the sphere. ┌…
ctx:discord/blah/watt-activation/500- full textwatt-activation-500text/plain3 KB
doc:agent/watt-activation-500/abfa0a8a-2310-4c13-88a1-8d09e2b08501Show excerpt
[2026-03-22 18:23] xenonfun: thankfully it seems <@806444151422976035> has prevented recusion from you talking to yourself. [2026-03-22 18:24] xenonfun: Working. The scalar oscillates (7.5 → 7.3), vector decays, bivector grows — the oscilla…
See also
- Kan Attention
- Channel Noise Bandwidth
- Phase Dependent Coupling
- Dyn H4 Phase Run
- Manifold Projection
- Activation Functions
- Coupling
- Sync
- Weights
- Linear Layers
- True
- Simulation
- Concept
- Dynamics
- Channel Noise
- Bandwidth
- Between Linear Layers
- Activation Function
- Mechanism
- Computational Method
- Sphere
- Scalar Oscillation 7.5 to 7.3
- Vector Decay
- Bivector Growth
- Commutator Limitation
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.