latest.pt
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
latest.pt has 27 facts recorded in Dontopedia across 9 references, with 2 live disagreements.
Mostly:points to(5), evaluated at step(2), rdf:type(2)
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.
concernsModelCheckpointConcerns Model Checkpoint(1)
- This Message
ex:this-message
criticizesRepetitionCriticizes Repetition(1)
- Xenonfun
ex:xenonfun
exhibitedByExhibited by(1)
- Proportion Construction Species Atomic Loops
ex:proportion-construction-species-atomic-loops
marksCheckpointMarks Checkpoint(1)
- Step 2000
ex:step-2000
observesCleanerBytesObserves Cleaner Bytes(1)
- Xenonfun
ex:xenonfun
performsBetterThanPerforms Better Than(1)
- Best Pt
ex:best-pt
remainsSuperiorToRemains Superior to(1)
- Rolling Best
ex:rolling-best
savedInCheckpointSaved in Checkpoint(1)
- Train Bpb
ex:train-bpb
superiorSemanticallySuperior Semantically(1)
- Best Pt
ex:best-pt
Other facts (26)
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 |
|---|---|---|
| Points to | Step 0004000 Pt | [1] |
| Points to | Step 0000250 Pt | [2] |
| Points to | Step 0001000 Pt | [3] |
| Points to | Step 0001000 Pt | [5] |
| Points to | Step 0004000 Pt | [8] |
| Evaluated at Step | 2000 | [4] |
| Evaluated at Step | 2000 | [6] |
| Rdf:type | File | [8] |
| Rdf:type | Checkpoint File | [9] |
| Worse on Holdout Than | Best Pt | [4] |
| Is Checkpoint at | Step 2000 | [4] |
| Has Validation Bpb | 1.3714 | [4] |
| Has Worse Semantics | Best Pt | [6] |
| Improved Byte Cleanliness | Best Pt | [6] |
| Compared to | Best Pt | [6] |
| Loses Topic | Nhl Topic | [6] |
| Exhibits Repetition Loops | proportion / construction / species / atomic | [6] |
| Generates Continuation | Continuation | [6] |
| Has Cleaner Bytes | true | [6] |
| Has Fewer | Nul Separators | [6] |
| Has Train Bpb | 1.3653 | [6] |
| Has Worse Repetition | Best Pt | [6] |
| Stuck in Loops | Proportion Construction Species Atomic Loops | [6] |
| Used Validation Prompt | Validation Prompt | [6] |
| Updated to | Step 0005000 Pt | [7] |
| Use Condition | final curve is healthy | [9] |
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 (9)
ctx:discord/blah/watt-activation/part-701ctx:discord/blah/watt-activation/part-705ctx:discord/blah/watt-activation/part-707ctx:discord/blah/watt-activation/part-712ctx:discord/blah/watt-activation/part-706ctx:discord/blah/watt-activation/part-713ctx:discord/blah/watt-activation/part-714ctx:discord/blah/watt-activation/698- full textwatt-activation-698text/plain3 KB
doc:agent/watt-activation-698/f8b43921-7946-4fa9-adc2-3852b2c781f2Show excerpt
[2026-05-01 04:58] xenonfun: ``` Checked it out. My take: the handoff is strong and mostly accurate, but there are two path mismatches worth fixing. e23 training is healthy and still running. Latest metric I saw was step `4350 / 25000`, ab…
ctx:discord/blah/watt-activation/706- full textwatt-activation-706text/plain2 KB
doc:agent/watt-activation-706/6633be5c-fab3-495f-b5b5-8dadec7f0fbbShow excerpt
[2026-05-01 23:01] xenonfun: Yes. The 5K gate run is launched and behaving well so far. Current run: `e24_chinchilla_latest_governed_5k` It is warm-started from e23 best with: - `B=20`, `T=2048` - updated Chinchilla dataset, `6,123,670` t…
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.