Dontopedia

Training Step 1500

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

Training Step 1500 has 22 facts recorded in Dontopedia across 5 references, with 1 live disagreement.

22 facts·14 predicates·5 sources·1 in dispute

Mostly:has loss(4), has step number(2), has best bpb(2)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound 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.

precedesPrecedes(1)

Other facts (22)

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.

22 facts
PredicateValueRef
Has Loss1.7411[1]
Has Loss1.7365[2]
Has Loss1.7715[3]
Has Loss1.7715[5]
Has Step Number1500[2]
Has Step Number1500[5]
Has Best Bpb2.453[3]
Has Best Bpb2.453[5]
Has Bpb2.556[3]
Has Bpb2.556[5]
Has Learning Rate0.003[3]
Has Learning Rate0.003[5]
Rdf:typeTraining Step[4]
Rdf:typeTraining Step[5]
Is Out of2000[1]
PrecedesTraining Step 2000[2]
Has Tokens Per Second5780[3]
Step Number1500[4]
Reported Loss1.7365[4]
Part ofTraining Log 2026 04 20 02 56[5]
Has Throughput Tok Per Sec5780[5]
Has Domain WeightsDoremi Step 1500 Weights[5]

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.

22 undated
—
hasLossblah/vidya/part-4
1.7411
—
isOutOfblah/vidya/part-4
2000
—
precedesblah/vidya/part-3
ex:training-step-2000
—
hasLossblah/vidya/part-3
1.7365
—
hasStepNumberblah/vidya/part-3
1500
—
hasBestBPBblah/watt-activation/part-666
2.453
—
hasBPBblah/watt-activation/part-666
2.556
—
hasLearningRateblah/watt-activation/part-666
0.003
—
hasLossblah/watt-activation/part-666
1.7715
—
hasTokensPerSecondblah/watt-activation/part-666
5780
—
typeblah/vidya/3
ex:TrainingStep
—
stepNumberblah/vidya/3
1500
—
reportedLossblah/vidya/3
1.7365
—
typeblah/watt-activation/663
ex:TrainingStep
—
partOfblah/watt-activation/663
ex:training-log-2026-04-20-02-56
—
hasStepNumberblah/watt-activation/663
1500
—
hasLossblah/watt-activation/663
1.7715
—
hasBPBblah/watt-activation/663
2.556
—
hasBestBPBblah/watt-activation/663
2.453
—
hasLearningRateblah/watt-activation/663
0.003
—
hasThroughputTokPerSecblah/watt-activation/663
5780
—
hasDomainWeightsblah/watt-activation/663
ex:doremi-step-1500-weights

References (5)

5 references
  1. [1]Part 42 facts
    ctx:discord/blah/vidya/part-4
  2. [2]Part 33 facts
    ctx:discord/blah/vidya/part-3
  3. [3]Part 6665 facts
    ctx:discord/blah/watt-activation/part-666
  4. [4]33 facts
    ctx:discord/blah/vidya/3
    • full textvidya-3
      text/plain2 KBdoc:agent/vidya-3/36dc77aa-7dff-487d-af09-d932dc181732
      Show excerpt
      [2026-02-20 07:37] rolandnsharp7643: roland@cube:~/code/flow/ocaml$ ./microgpt_scaled num docs: 16478 vocab size: 80 num params: 206848 step 500 / 2000 | loss 2.5678 step 1000 / 2000 | loss 2.1392 step 1500 / 2000 | loss 1.7365 step 2000 /…
  5. [5]6639 facts
    ctx:discord/blah/watt-activation/663
    • full textwatt-activation-663
      text/plain3 KBdoc:agent/watt-activation-663/c85e6460-7efe-4554-a28a-50b22ebd478d
      Show excerpt
      [2026-04-20 02:51] xenonfun: ``` ⏺ Bench results (single-layer, 25M shape G=7 d_osc=74 d_model=518): ┌─────┬────────┬────────┬───────┐ │ T │ fwd ms │ bwd ms │ ratio │ ├─────┼────────┼────────┼───────┤ │ 64 │ 0.149 │ 0.218 │ 1.…

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.

↑