Decoder Confidence
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Decoder Confidence has 8 facts recorded in Dontopedia across 3 references.
Mostly:shows bytes model attends to(1), plots predicted logits peak(1), recommended for usefulness(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (1)
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.
shouldShowOneOfShould Show One of(1)
- Panel
ex:panel
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.
| Predicate | Value | Ref |
|---|---|---|
| Shows Bytes Model Attends to | Bytes | [1] |
| Plots Predicted Logits Peak | Constellation | [1] |
| Recommended for Usefulness | null | [1] |
| Per Current Step | Output Closeness | [2] |
| Measures Closeness to | Constellation Points | [2] |
| Rdf:type | Visualization Feature | [3] |
| Metric Definition | how close the model's output is to each constellation point | [3] |
| Measurement Scope | for the current step | [3] |
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-332ctx:discord/blah/watt-activation/part-329ctx:discord/blah/watt-activation/327- full textwatt-activation-327text/plain2 KB
doc:agent/watt-activation-327/2321ad6e-6bc7-4a98-ab9f-4e2b23088910Show excerpt
[2026-03-15 04:28] xenonfun: That's remarkable consistency — 0.001 BPB variance across seeds, within 2 seconds of runtime. The architecture is stable, the result is reproducible, and the slope is slightly steeper on seed43 which is a good s…
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.