CNN Benchmark

as donto believed on

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

CNN Benchmark has 49 facts recorded in Dontopedia across 11 references, with 5 live disagreements.

49 facts·32 predicates·11 sources·5 in dispute

Mostly:has result row(9), compares(4), involves(3)

Inbound 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.

`canRun`Can Run(1)

Symbiogenesisex:symbiogenesis

`comparesAgainst`Compares Against(1)

Cross Couple Training Runex:cross-couple-training-run

`hasExperiment`Has Experiment(1)

Symbiogenesis Replication Studyex:symbiogenesis-replication-study

`include`Include(1)

Four Experimentsex:four-experiments

`isTrainCodeLmOnVqOutputs`Is Train Code Lm on Vq Outputs(1)

Experiment 1ex:experiment-1

`notConfirmedBy`Not Confirmed by(1)

Low Depth Spectral Coverage Hypothesisex:low-depth-spectral-coverage-hypothesis

Other facts (48)

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.

48 facts

Predicate	Value	Ref
Has Result Row	Result Row 1	[5]
Has Result Row	Result Row 2	[5]
Has Result Row	Result Row 3	[5]
Has Result Row	Result Row 4	[5]
Has Result Row	Result Row 5	[5]
Has Result Row	Result Row 6	[5]
Has Result Row	Result Row 7	[5]
Has Result Row	Result Row 8	[5]
Has Result Row	Result Row 9	[5]
Compares	V3 Model	[3]
Compares	Helmholtz Model	[3]
Compares	Symbiogenesis	[11]
Compares	Baselines	[11]
Involves	Mnist Dataset 5k 1k	[11]
Involves	Fashion Mnist Dataset 1k 500	[11]
Involves	Cifar 10 Dataset 2k 1k	[11]
Has Test Run	Test Run Adam	[6]
Has Test Run	Test Run Fast Muon	[6]
Rdf:type	Experiment	[8]
Rdf:type	Experiment	[10]
Involves Model	Helmholtz	[9]
Involves Model	V3	[9]
Includes Concurrent Run Thru at Start	true	[1]
Partially Complete	true	[1]
Might Need Fresh Run	true	[1]
Is Train Code Lm on Vq Outputs	Experiment 1	[2]
Named	Depth Efficiency — Complete	[3]
Falsifies	Helmholtz Graceful Degradation Hypothesis	[3]
Used Venom Amount	1/4 gr	[4]
Killed Guinea Pig in	30 minutes	[4]
References Table	Paper Table 1	[5]
Has Configuration	iterations=60	[5]
Uses Starting Checkpoint	Checkpoint 9198	[6]
Has Baseline Metric	Baseline Metric 1	[6]
Has Name	anchor_kan+lohe_v3	[7]
Has Status	running	[7]
Has Step Number	58	[7]
Has Data Stage	early	[7]
Estimated Completion Time	20 min	[7]
Has Key Findings	Key Findings Set	[7]
Prompt Count	5	[8]
Outcome	same structure	[8]
Resulting State	nearly identical outputs	[8]
Classified As	baseline	[8]
Has Title	Exp 1: Depth efficiency — H vs v3 at 6/4/3 layers × 3 seeds	[9]
Has Layer Configurations	6/4/3 layers	[9]
Has Seed Count	3	[9]
Is Type of	Cnn Benchmark	[11]

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.

49 undated

—

includesConcurrentRunThruAtStartblah/watt-activation/part-1

true

—

partiallyCompleteblah/watt-activation/part-1

true

—

mightNeedFreshRunblah/watt-activation/part-1

true

—

isTrainCodeLmOnVqOutputsblah/watt-activation/part-288

ex:experiment-1

—

namedblah/watt-activation/part-455

Depth Efficiency — Complete

—

comparesblah/watt-activation/part-455

ex:v3-model

—

comparesblah/watt-activation/part-455

ex:helmholtz-model

—

falsifiesblah/watt-activation/part-455

ex:helmholtz-graceful-degradation-hypothesis

—

usedVenomAmounttrove-cooktown/chinese-fishermen

1/4 gr

—

killedGuineaPigIntrove-cooktown/chinese-fishermen

30 minutes

—

labelblah/watt-activation/1

CNN Benchmark

—

referencesTableblah/watt-activation/1

ex:paper-table-1

—