R Metric
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
R Metric has 17 facts recorded in Dontopedia across 11 references, with 1 live disagreement.
Mostly:measures coherence(2), increases over training(1), lower is better(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (9)
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.
avoidsOverSyncAvoids Over Sync(1)
- Training Diagnostics 1
ex:training-diagnostics-1
causesRIncreaseCauses R Increase(1)
- Higher Lr
ex:higher-lr
correlatesWithLowerRCorrelates With Lower R(1)
- Healthier Model
ex:healthier-model
exhibitsLowCoherenceExhibits Low Coherence(1)
- Training Diagnostics 1
ex:training-diagnostics-1
extendsBeyondRPerGroupExtends Beyond R Per Group(1)
- Inter Group Phase Coherence
ex:inter-group-phase-coherence
hasVeryLowCoherenceHas Very Low Coherence(1)
- G8 H2 Bpsk16 Helmholtz
ex:g8-h2-bpsk16-helmholtz
includesIncludes(1)
- Logged Per Step Per Block
ex:logged-per-step-per-block
includesMetricIncludes Metric(1)
- Current Measurements
ex:current-measurements
presupposesRAsSynchronizationMeasurePresupposes R As Synchronization Measure(1)
- R Adaptive Lr
ex:r-adaptive-lr
Other facts (17)
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 |
|---|---|---|
| Measures Coherence | true | [3] |
| Measures Coherence | Models | [7] |
| Increases Over Training | null | [1] |
| Lower Is Better | {} | [2] |
| Has Structured Regime Threshold | <0.85 | [4] |
| Considered Diagnostic | Model Performance | [5] |
| Implies Synchronization Level | null | [5] |
| Is Synchronization Metric | null | [5] |
| Peaks Around Step100 | null | [6] |
| Current Value | 0.31 | [8] |
| Previous Value | 0.27 | [8] |
| Increases Due to Lr | Higher Lr | [8] |
| Lower Is Healthier | Model Health | [9] |
| Analogous to Mean Field Loudness | how loud the mean field is | [10] |
| Measures Fundamental Resonance Strength | true | [10] |
| Named | r | [10] |
| Rdf:type | Metric | [11] |
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 (11)
ctx:discord/blah/watt-activation/part-49ctx:discord/blah/watt-activation/part-52ctx:discord/blah/watt-activation/part-186ctx:discord/blah/watt-activation/part-316ctx:discord/blah/watt-activation/part-393ctx:discord/blah/watt-activation/part-386ctx:discord/blah/watt-activation/part-395ctx:discord/blah/watt-activation/part-398ctx:discord/blah/watt-activation/part-404ctx:discord/blah/watt-activation/part-414ctx:discord/blah/watt-activation/488- full textwatt-activation-488text/plain3 KB
doc:agent/watt-activation-488/3b1655f5-5846-4949-99fa-94517d956b67Show excerpt
[2026-03-22 04:58] xenonfun: ⏺ 12.8K tok/s! Up from 8.6K with cached norms alone. The NEON SIMD dot_slice with vfmaq_f32 (fused multiply-add on 4 floats) gives another 49% speedup. Total speedup from the original 5.7K: ``` ┌───────────…
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.