Dontopedia

Step 1000

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

Step 1000 has 65 facts recorded in Dontopedia across 13 references, with 9 live disagreements.

65 facts·44 predicates·13 sources·9 in dispute

Mostly:has loss(7), rdf:type(4), has lr(3)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (3)

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)

reachedStepReached Step(1)

savedFromStepSaved From Step(1)

Other facts (64)

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.

64 facts
PredicateValueRef
Has Loss2.2229[1]
Has Loss2.1392[2]
Has Loss5.4467[3]
Has Loss6.0333[4]
Has Loss1.8779[6]
Has Loss1.2008[7]
Has Loss2.875[11]
Rdf:typeTraining Step[8]
Rdf:typeTraining Event[9]
Rdf:typeTraining Step[12]
Rdf:typeTraining Step[13]
Has Lr0.0000996[4]
Has Lr0.003[6]
Has Lr0.002[7]
Has Bpb2.709[6]
Has Bpb1.732[7]
Has Bpb4.15[11]
Step Number1000[8]
Step Number1000[12]
Step Number1000[13]
Has Tokens Per Second12285[3]
Has Tokens Per Second10653[12]
Has Learning Rate1.99e-05[3]
Has Learning Rate0.002[12]
Has Tok Per S77537[4]
Has Tok Per S10653[7]
Throughput UnitB/s[10]
Throughput Unittok/s[13]
Has Throughput110,369-B/s[11]
Has Throughput13891[13]
Is Out of2000[1]
Has Step Number1000[2]
PrecedesTraining Step 1500[2]
Has Eta244min[3]
Has Perplexity232[3]
Achieves Best Validation So FarVal Loss 5 4712[3]
Occurs Before Total Completion22920[3]
Has Current Step1000[3]
Has Epochep02/30[3]
Has Total Steps22920[3]
Has Step Time667ms[3]
Has Progress Percentage30.8[3]
Precedes ValidationValidation Step 1000[4]
Progress Percent8[4]
Has Eta Min40[4]
Has Ppl417.1[4]
Has Step Time Ms211[4]
S1 Value0.027[5]
Bpb Value2.901[5]
S2 Value0.014[5]
Has Eval Bpb2.763[6]
Has Tok Per Sec7316[6]
Has Best Bpb2.607[6]
Has Gnorm0.6481[7]
Reported Loss2.1392[8]
Perplexity9[10]
Utilization0.234[10]
Throughput197267[10]
Part ofCross Patch Ar Decoder[11]
Has Loss Value1.2008[12]
Has Bpb Value1.732[12]
Has Gradient Norm0.6481[12]
Has Loss Metric0.9332[13]
Has Bpb Metric1.3463[13]

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.

65 undated
—
isOutOfblah/vidya/part-4
2000
—
hasLossblah/vidya/part-4
2.2229
—
hasStepNumberblah/vidya/part-3
1000
—
hasLossblah/vidya/part-3
2.1392
—
precedesblah/vidya/part-3
ex:training-step-1500
—
hasEtablah/watt-activation/part-165
244min
—
hasPerplexityblah/watt-activation/part-165
232
—
achievesBestValidationSoFarblah/watt-activation/part-165
ex:val-loss-5-4712
—
occursBeforeTotalCompletionblah/watt-activation/part-165
22920
—
hasCurrentStepblah/watt-activation/part-165
1000
—
hasEpochblah/watt-activation/part-165
ep02/30
—
hasTotalStepsblah/watt-activation/part-165
22920
—
hasTokensPerSecondblah/watt-activation/part-165
12285
—
hasLearningRateblah/watt-activation/part-165
1.99e-05
—
hasLossblah/watt-activation/part-165
5.4467
—
hasStepTimeblah/watt-activation/part-165
667ms
—
hasProgressPercentageblah/watt-activation/part-165
30.8
—
precedesValidationblah/watt-activation/part-238
ex:validation-step-1000
—
progressPercentblah/watt-activation/part-238
8
—
hasEtaMinblah/watt-activation/part-238
40
—
hasLrblah/watt-activation/part-238
0.0000996
—
hasTokPerSblah/watt-activation/part-238
77537
—
hasLossblah/watt-activation/part-238
6.0333
—
hasPplblah/watt-activation/part-238
417.1
—
hasStepTimeMsblah/watt-activation/part-238
211
—
s1Valueblah/watt-activation/part-372
0.027
—
bpbValueblah/watt-activation/part-372
2.901
—
s2Valueblah/watt-activation/part-372
0.014
—
hasEvalBpbblah/watt-activation/part-644
2.763
—
hasTokPerSecblah/watt-activation/part-644
7316
—
hasBestBpbblah/watt-activation/part-644
2.607
—
hasBpbblah/watt-activation/part-644
2.709
—
hasLrblah/watt-activation/part-644
0.003
—
hasLossblah/watt-activation/part-644
1.8779
—
hasLrblah/watt-activation/part-658
0.002
—
hasLossblah/watt-activation/part-658
1.2008
—
hasGnormblah/watt-activation/part-658
0.6481
—
hasTokPerSblah/watt-activation/part-658
10653
—
hasBpbblah/watt-activation/part-658
1.732
—
typeblah/vidya/3
ex:TrainingStep
—
stepNumberblah/vidya/3
1000
—
reportedLossblah/vidya/3
2.1392
—
typeblah/watt-activation/297
ex:TrainingEvent
—
perplexityblah/watt-activation/299
9
—
utilizationblah/watt-activation/299
0.234
—
throughputblah/watt-activation/299
197267
—
throughputUnitblah/watt-activation/299
B/s
—
hasLossblah/watt-activation/300
2.875
—
hasBpbblah/watt-activation/300
4.15
—
hasThroughputblah/watt-activation/300
110,369-B/s
—
partOfblah/watt-activation/300
ex:cross-patch-ar-decoder
—
typeblah/watt-activation/655
ex:TrainingStep
—
stepNumberblah/watt-activation/655
1000
—
hasLossValueblah/watt-activation/655
1.2008
—
hasBpbValueblah/watt-activation/655
1.732
—
hasGradientNormblah/watt-activation/655
0.6481
—
hasLearningRateblah/watt-activation/655
0.002
—
hasTokensPerSecondblah/watt-activation/655
10653
—
labelblah/watt-activation/703
Step 1000
—
typeblah/watt-activation/703
ex:TrainingStep
—
stepNumberblah/watt-activation/703
1000
—
hasLossMetricblah/watt-activation/703
0.9332
—
hasBpbMetricblah/watt-activation/703
1.3463
—
hasThroughputblah/watt-activation/703
13891
—
throughputUnitblah/watt-activation/703
tok/s

References (13)

13 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 16512 facts
    ctx:discord/blah/watt-activation/part-165
  4. [4]Part 2388 facts
    ctx:discord/blah/watt-activation/part-238
  5. [5]Part 3723 facts
    ctx:discord/blah/watt-activation/part-372
  6. [6]Part 6446 facts
    ctx:discord/blah/watt-activation/part-644
  7. [7]Part 6585 facts
    ctx:discord/blah/watt-activation/part-658
  8. [8]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 /…
  9. [9]2971 fact
    ctx:discord/blah/watt-activation/297
    • full textwatt-activation-297
      text/plain2 KBdoc:agent/watt-activation-297/ad91f718-f038-464f-a6d2-ba91d77fe4e3
      Show excerpt
      [2026-03-14 05:23] xenonfun: 600K context, the UI crazy scroll thing and memory leaks are annoyin claude is sucking up 6GB now. [2026-03-14 05:24] xenonfun: ``` ⏺ Launched in rjs:longgen. This trains with longer context (512) and more steps…
  10. [10]2994 facts
    ctx:discord/blah/watt-activation/299
    • full textwatt-activation-299
      text/plain2 KBdoc:agent/watt-activation-299/3c2f94b1-841d-457c-8bfa-1161d5e80363
      Show excerpt
      [2026-03-14 06:27] xenonfun: ``` Loaded 41.4 MB VQ encoder (5000 steps)... Frozen downstream: 170,329 trainable / 520,945 total step 1000 | ppl=9.0 util=0.234 | 197,267 B/s step 2000 | ppl=31.0 util=0.594 | 196,457 B/s st…
  11. [11]3004 facts
    ctx:discord/blah/watt-activation/300
    • full textwatt-activation-300
      text/plain3 KBdoc:agent/watt-activation-300/3b6edccf-3524-4608-838f-25890efaea15
      Show excerpt
      [2026-03-14 06:34] xenonfun: ``` 3. Manual attention (lines 110-128) — Hand-rolled softmax attention instead of using mx.fast.scaled_dot_product_attention. MLX's fused attention kernel is significantly faster for small sequence lengths. …
  12. [12]6557 facts
    ctx:discord/blah/watt-activation/655
    • full textwatt-activation-655
      text/plain2 KBdoc:agent/watt-activation-655/184fbb55-c5df-4c57-ba7d-a525dd418a7a
      Show excerpt
      [2026-04-19 09:46] xenonfun: loaded 24999946 bytes from /Users/ms/MS/HarmonicRust/data/curriculum/domain_tinystories.bin step 0 loss=6.1092 bpb=8.814 gnorm=10.6885 lr=4.00e-6 9259 tok/s step 50 loss=2.8955 bpb=4.177 gnorm=2.2132…
  13. [13]7037 facts
    ctx:discord/blah/watt-activation/703
    • full textwatt-activation-703
      text/plain3 KBdoc:agent/watt-activation-703/c2aff1b4-7f23-419c-b531-687b267a9440
      Show excerpt
      [2026-05-01 20:40] xenonfun: 1K smoke finished and inference is done. Run: `/home/ms/data/clifford_ssm/checkpoints/e24_chinchilla_from_e23_best_prefetch_1k/latest.pt` Final training stats: - step `1000` - loss / BPB: `0.9332 / 1.3463` - 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.

↑