Dontopedia

Benchmark table

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

Benchmark table has 11 facts recorded in Dontopedia across 6 references, with 2 live disagreements.

11 facts·7 predicates·6 sources·2 in dispute

Mostly:contains config(3), demonstrates scaling(1), compares cpu before and gpu(1)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (2)

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.

containsDataContains Data(1)

presentsTableComparisonPresents Table Comparison(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.

9 facts
PredicateValueRef
Contains ConfigConfig M32 L512[4]
Contains ConfigConfig M32 L1024[4]
Contains ConfigConfig M32 L2048[4]
Demonstrates Scaling{}[1]
Compares Cpu Before and Gpu{}[1]
Emphasizes DriftZero Percent[2]
DeviceMps[3]
Rdf:typeTable[4]
Describes SubjectChon NASA Ims Bearing[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.

11 undated
—
demonstratesScalingblah/watt-activation/part-528
{}
—
comparesCpuBeforeAndGpublah/watt-activation/part-528
{}
—
emphasizesDriftblah/watt-activation/part-566
ex:zero-percent
—
deviceblah/watt-activation/8
ex:mps
—
typeblah/watt-activation/75
ex:Table
—
containsConfigblah/watt-activation/75
ex:config-m32-l512
—
containsConfigblah/watt-activation/75
ex:config-m32-l1024
—
containsConfigblah/watt-activation/75
ex:config-m32-l2048
—
labelblah/watt-activation/512
Final benchmark table for this session
—
describesSubjectblah/watt-activation/512
ex:chon-nasa-ims-bearing
—
labelblah/watt-activation/654
Benchmark table

References (6)

6 references
  1. [1]Part 5282 facts
    ctx:discord/blah/watt-activation/part-528
  2. [2]Part 5661 fact
    ctx:discord/blah/watt-activation/part-566
  3. [3]81 fact
    ctx:discord/blah/watt-activation/8
    • full textwatt-activation-8
      text/plain3 KBdoc:agent/watt-activation-8/c065f0d3-ec70-4dc8-b305-5fd7a4e6d77a
      Show excerpt
      [2026-02-27 01:08] xenonfun: ``` ### implement phase-gated residual (#5): x + r * h instead of x + h in KuramotoBlock This is great data. At epoch 9 it's already at PPL 86.11 — matching the no-FFN baseline's final PPL of 86.09, but: - n…
  4. [4]754 facts
    ctx:discord/blah/watt-activation/75
    • full textwatt-activation-75
      text/plain2 KBdoc:agent/watt-activation-75/24e73a49-4879-4ec1-9e4e-461a2fd9d66e
      Show excerpt
      [2026-03-07 18:23] xenonfun: Old per-anchor-loop benchmark complete. Key observations: 1. PPL parity confirmed — aKAN matches KAN within 0.3 PPL at every sequence length 2. Anchor entropy: H=3.46, eff_anc=31.8/32 — at random init / 2K …
  5. [5]5122 facts
    ctx:discord/blah/watt-activation/512
    • full textwatt-activation-512
      text/plain2 KBdoc:agent/watt-activation-512/b9562690-d0ae-4a31-b0ba-f7ce99f7c320
      Show excerpt
      [2026-03-22 21:20] xenonfun: ⏺ MAE 9.77% — same as plain MSE (9.8%). The weighting doesn't hurt but doesn't help either for this dataset. The early-life predictions are already good because the CHON features naturally separate healthy fr…
  6. [6]6541 fact
    ctx:discord/blah/watt-activation/654
    • full textwatt-activation-654
      text/plain3 KBdoc:agent/watt-activation-654/64b35524-e5cb-4a0b-b1df-4ce474f68552
      Show excerpt
      [2026-04-18 07:57] xenonfun: ``` Standalone signing — this is a much better story NA-PDE-Sig Cl(5,0) grade-even is structurally different from the DH leg: - Different hardness assumption. Non-abelian conjugacy in Cl⁺(5,0) ≅ M₄(F_p) — …

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.

↑