Dontopedia

Current trained weights

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

Current trained weights has 9 facts recorded in Dontopedia across 2 references.

9 facts·8 predicates·2 sources

Mostly:performs no better than chance(1), has percentage near ternary values after scaling(1), are not(1)

Maturity scale raw canonical shape-checked rule-derived certified

Other facts (8)

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.

8 facts
PredicateValueRef
Performs No Better Than ChanceRandom Gaussian[1]
Has Percentage Near Ternary Values After Scaling~46%[1]
Are NotNaturally Ternary[1]
Presupposes Existence ofLohe Model[1]
Rdf:typeModel Parameter[2]
Is Not Naturally Ternarytrue[2]
Performs Better Than Chancefalse[2]
Performance Compared toRandom Gaussian Baseline[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.

performsNoBetterThanChanceblah/watt-activation/part-277
ex:random-gaussian
hasPercentageNearTernaryValuesAfterScalingblah/watt-activation/part-277
~46%
areNotblah/watt-activation/part-277
ex:naturally-ternary
presupposesExistenceOfblah/watt-activation/part-277
ex:lohe-model
typeblah/watt-activation/275
ex:ModelParameter
labelblah/watt-activation/275
Current trained weights
isNotNaturallyTernaryblah/watt-activation/275
true
performsBetterThanChanceblah/watt-activation/275
false
performanceComparedToblah/watt-activation/275
ex:random-gaussian-baseline

References (2)

2 references
  1. [1]Part 2774 facts
    ctx:discord/blah/watt-activation/part-277
  2. [2]2755 facts
    ctx:discord/blah/watt-activation/275
    • full textwatt-activation-275
      text/plain3 KBdoc:agent/watt-activation-275/149bf24b-54fb-4412-b289-e2e03bccffe2
      Show excerpt
      [2026-03-13 20:46] xenonfun: ⏺ This tells a clear story: Current trained weights are NOT naturally ternary. Only ~46% of values are near {-1, 0, 1} after scaling (random Gaussian would give ~47%, so no better than chance). Relative err

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.