From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
V8k Training has 60 facts recorded in Dontopedia across 2 references, with 3 live disagreements.
Mostly:is live(2), variance reason(2), has loss at iteration(2)
totalIterationscurrentIterationavgLossAtIter4500comparedToV2kModelcurrentLRembeddingsNeedIterationsForPhilosophyVocabetaCompletionTimeOther 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.
providesStatusUpdateProvides Status Update(1)ex:xenonfunThe 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 |
|---|---|---|
| Eta Minutes | 92 | [2] |
| Expected Catchup Iterations | 20-30K | [2] |
| Expects High Variance Per Step | true | [2] |
| Gap With V7 at Iter4k | Output Quality Gap | [2] |
| Has Attachment | Screenshot 2026 03 06 at 4 18 01 Pm Png | [2] |
| Has Progress Percent | 0.8 | [2] |
| Has Speed It Per S | 17.1 | [2] |
| Has Tok Per S | 4400 | [2] |
| Is Fine Tuning of | Different Philosophy | [2] |
| Is Vocab Expansion Trial | Vocab 8k Expansion | [2] |
| Lr Peak Iteration | 10000 | [2] |
| Lr Peak Value | 0.0003 | [2] |
| Produces Sports Wiki Noise at Iter4k | true | [2] |
| Progress Percentage | 4.5 | [2] |
| Remaining Time At21 33 | 90 | [2] |
| Speed It Per S | 17.3 | [2] |
| Starting Avg Loss at Iter100 | 7.8 | [2] |
| Uses Tokenized Corpus | 8k Tokenized Corpus | [2] |
| Warmup Phase Length | 10000 | [2] |
| Catches Up to V2k After | 20 30k Iters | [2] |
| Predicted to Improve | Philosophy Vocab | [2] |
| Presupposes Ongoing | true | [2] |
| Has Variance Property | High variance per-step | [1] |
| Has Expected Loss Property | familiar token space | [1] |
| Has Comparative Model | 2k Model | [1] |
| Session Name | train_v8k | [1] |
| Estimated Finish Time Local | 17:00 | [1] |
| Estimated Time Remaining | 5520 | [1] |
| Expected Catch Up Iteration | 25000 | [1] |
| Uses Corpus | 8k Tokenized Corpus | [1] |
| Peak Learning Rate | 0.0003 | [1] |
| Warmup Iterations | 10000 | [1] |
| In Warmup Phase | true | [1] |
| Learning Rate | 0.000151 | [1] |
| Loss Trend | dropping steadily | [1] |
| Average Loss | 5.71 | [1] |
| Processing Speed Tokens | 4400 | [1] |
| Processing Speed Iterations | 17.3 | [1] |
| Iteration Percentage | 4.5 | [1] |
| Has Status | healthy | [1] |
| Rdfs:label | v8k training | [1] |
| Rdf:type | Training Run | [1] |
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.
doc:agent/watt-activation-31/46160fba-f33d-4eec-8b0a-faef1e38bef4[2026-03-06 21:22] xenonfun: Vocab 8K expansion trial. much slower ` 17.1 it/s (4.4K tok/s | 0.80%)` Training is live and healthy. Here's the status: v8k training — iter 4500 / 100000 (4.5%) - Speed: 17.3 it/s, ~92 min remaining - Lo…
Dontopedia is in a read-only public launch. Follow the references and disputed branches now; contributions will open after durable identity and moderation are in place.