softmax
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
softmax has 13 facts recorded in Dontopedia across 3 references.
Mostly:violates design principles(1), assumes competition(1), exception condition(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (6)
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.
assumedByAssumed by(1)
- Euclidean Logit Competition
ex:euclidean-logit-competition
makesSoftmaxUnnecessaryMakes Softmax Unnecessary(1)
- 9 Token Causal Sequence
ex:9-token-causal-sequence
mentionsArchitectureMentions Architecture(1)
- Message 2026 03 12 16 31
ex:message-2026-03-12-16-31
outperformsOutperforms(1)
- Kuramoto Attention
ex:kuramoto-attention
rendersUnnecessaryRenders Unnecessary(1)
- 9 Token Causal Sequence
ex:9-token-causal-sequence
usesAttentionTypeUses Attention Type(1)
- Lohegpt V5
ex:lohegpt-v5
Other facts (12)
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 |
|---|---|---|
| Violates Design Principles | Architecture | [1] |
| Assumes Competition | Euclidean Logit Competition | [1] |
| Exception Condition | mathematically unavoidable | [1] |
| Incompatible Ontologically | Lohe Dynamics | [1] |
| Not Compatible With | Spherical Alignment | [1] |
| Presupposes Euclidean Assumption | Logit Competition | [1] |
| Provides No Functional Advantage | 9 Token Causal Sequence | [1] |
| Should Not Appear | Ar Decoder | [1] |
| Usage Condition | mathematically unavoidable | [2] |
| Assumes | Euclidean Logit Competition | [2] |
| Not Assumes | Spherical Alignment | [2] |
| Provides | no-functional-advantage | [2] |
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 (3)
ctx:discord/blah/watt-activation/part-302ctx:discord/blah/watt-activation/300- full textwatt-activation-300text/plain3 KB
doc:agent/watt-activation-300/3b6edccf-3524-4608-838f-25890efaea15Show excerpt
[2026-03-14 06:34] xenonfun: ``` 3. Manual attention (lines 110-128) — Hand-rolled softmax attention instead of using mx.fast.scaled_dot_product_attention. MLX's fused attention kernel is significantly faster for small sequence lengths. …
ctx:discord/blah/watt-activation/326- full textwatt-activation-326text/plain3 KB
doc:agent/watt-activation-326/13bd929c-8a9b-493f-b61c-5db579a2f719Show excerpt
[2026-03-15 04:22] xenonfun: # What The Literature Review Means For This Result The timing is significant. The search found that: - AKOrN (ICLR 2025) validated Kuramoto oscillators in transformers - SSA (February 2026) showed Kuramoto atte…
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.