Dontopedia

Step 300

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

Step 300 has 69 facts recorded in Dontopedia across 11 references, with 9 live disagreements.

69 facts·43 predicates·11 sources·9 in dispute

Mostly:has loss(10), has ppl(6), has elapsed time(4)

Maturity scale raw canonical shape-checked rule-derived certified

Has Lossin disputehasLoss

  • 9.6110[1]all time · Part 125
  • 18.0934[2]all time · Part 189
  • 11.5419[3]all time · Part 210
  • 9.0236[4]all time · Part 202
  • 8.8339[5]all time · Part 211
  • 9.0404[6]all time · Part 203
  • 9.16[7]all time · Part 228
  • 7.9793[8]all time · Part 216
  • 2.8361[9]all time · Part 331
  • 9.1581[11]sourceall time · 214

Inbound mentions (6)

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(3)

hasStepHas Step(1)

isStableVersionOfIs Stable Version of(1)

stabilizesByStepStabilizes by Step(1)

Other facts (58)

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.

58 facts
PredicateValueRef
Has Ppl14927.9[1]
Has Ppl72086414.6[2]
Has Ppl102939.2[3]
Has Ppl8296.8[4]
Has Ppl6863.2[5]
Has Ppl2919.9[8]
Has Elapsed Time154s[4]
Has Elapsed Time124s[5]
Has Elapsed Time204s[8]
Has Elapsed Time88[11]
Has Tok Per S17266[2]
Has Tok Per S13118[3]
Has Tok Per S8640[4]
Has Tokens Per Second12094[5]
Has Tokens Per Second107631[9]
Has Tokens Per Second14711[11]
Has Attn R0.6478[4]
Has Attn R0.6322[6]
Has Tok Per Sec15059[6]
Has Tok Per Sec6301[8]
Has Peak DifferentiationB9 Disorder Gap[7]
Has Peak DifferentiationSystem[7]
Has Perplexity17[9]
Has Perplexity9490.9[11]
Step Time630ms[1]
Eta172min[1]
Has Lr6.00e-05[1]
Tok Per S13001[1]
Progress1.8%[1]
Elapsed78[2]
Has Kappa0.177[3]
Has Time97s[3]
Has Sys R0.348[3]
Has Step Progress300/1000[4]
Has Overhead+127ms[4]
Has Extra Time+38ms[6]
Has Ppl8437[6]
Has Elapsed Secs85[6]
Has B10 Minus B110.037[7]
Has B10 R0.461[7]
Has B11 R0.424[7]
Has B9 Disorder0.131[7]
Has B9 R0.33[7]
PrecedesStep 1000[7]
Has Learning Rate1.80e-04[9]
Has Eta50min[9]
Is at Progress1.5%[9]
Has Rollout Steps0.12[9]
Has Ms Per Step152[9]
Rdf:typeTraining Step[10]
Loss Value9.0404[10]
Perplexity Value8437[10]
Tokens Per Second15059[10]
Elapsed Time85s[10]
Attn R Value0.6322[10]
Step Overhead+38ms[10]
Step Number300[11]
Part ofTraining Run 20260311 001319[11]

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.

stepTimeblah/watt-activation/part-125
630ms
etablah/watt-activation/part-125
172min
hasLrblah/watt-activation/part-125
6.00e-05
hasPplblah/watt-activation/part-125
14927.9
tokPerSblah/watt-activation/part-125
13001
hasLossblah/watt-activation/part-125
9.6110
progressblah/watt-activation/part-125
1.8%
elapsedblah/watt-activation/part-189
78
hasPplblah/watt-activation/part-189
72086414.6
hasLossblah/watt-activation/part-189
18.0934
hasTokPerSblah/watt-activation/part-189
17266
hasKappablah/watt-activation/part-210
0.177
hasTimeblah/watt-activation/part-210
97s
hasPplblah/watt-activation/part-210
102939.2
hasSysRblah/watt-activation/part-210
0.348
hasLossblah/watt-activation/part-210
11.5419
hasTokPerSblah/watt-activation/part-210
13118
hasLossblah/watt-activation/part-202
9.0236
hasStepProgressblah/watt-activation/part-202
300/1000
hasPplblah/watt-activation/part-202
8296.8
hasAttnRblah/watt-activation/part-202
0.6478
hasElapsedTimeblah/watt-activation/part-202
154s
hasTokPerSblah/watt-activation/part-202
8640
hasOverheadblah/watt-activation/part-202
+127ms
hasLossblah/watt-activation/part-211
8.8339
hasPplblah/watt-activation/part-211
6863.2
hasElapsedTimeblah/watt-activation/part-211
124s
hasTokensPerSecondblah/watt-activation/part-211
12094
hasAttnRblah/watt-activation/part-203
0.6322
hasTokPerSecblah/watt-activation/part-203
15059
hasExtraTimeblah/watt-activation/part-203
+38ms
hasLossblah/watt-activation/part-203
9.0404
hasPPLblah/watt-activation/part-203
8437
hasElapsedSecsblah/watt-activation/part-203
85
hasLossblah/watt-activation/part-228
9.16
hasB10MinusB11blah/watt-activation/part-228
0.037
hasB10Rblah/watt-activation/part-228
0.461
hasB11Rblah/watt-activation/part-228
0.424
hasB9Disorderblah/watt-activation/part-228
0.131
hasB9Rblah/watt-activation/part-228
0.33
hasPeakDifferentiationblah/watt-activation/part-228
ex:b9-disorder-gap
hasPeakDifferentiationblah/watt-activation/part-228
ex:system
precedesblah/watt-activation/part-228
ex:step-1000
hasLossblah/watt-activation/part-216
7.9793
hasTokPerSecblah/watt-activation/part-216
6301
hasPplblah/watt-activation/part-216
2919.9
hasElapsedTimeblah/watt-activation/part-216
204s
hasLossblah/watt-activation/part-331
2.8361
hasTokensPerSecondblah/watt-activation/part-331
107631
hasLearningRateblah/watt-activation/part-331
1.80e-04
hasEtablah/watt-activation/part-331
50min
isAtProgressblah/watt-activation/part-331
1.5%
hasRolloutStepsblah/watt-activation/part-331
0.12
hasPerplexityblah/watt-activation/part-331
17
hasMsPerStepblah/watt-activation/part-331
152
typeblah/watt-activation/202
ex:TrainingStep
lossValueblah/watt-activation/202
9.0404
perplexityValueblah/watt-activation/202
8437
tokensPerSecondblah/watt-activation/202
15059
elapsedTimeblah/watt-activation/202
85s
attnRValueblah/watt-activation/202
0.6322
stepOverheadblah/watt-activation/202
+38ms
labelblah/watt-activation/214
Step 300
stepNumberblah/watt-activation/214
300
partOfblah/watt-activation/214
ex:training-run-20260311-001319
hasLossblah/watt-activation/214
9.1581
hasPerplexityblah/watt-activation/214
9490.9
hasTokensPerSecondblah/watt-activation/214
14711
hasElapsedTimeblah/watt-activation/214
88

References (11)

11 references
  1. [1]Part 1257 facts
    ctx:discord/blah/watt-activation/part-125
  2. [2]Part 1894 facts
    ctx:discord/blah/watt-activation/part-189
  3. [3]Part 2106 facts
    ctx:discord/blah/watt-activation/part-210
  4. [4]Part 2027 facts
    ctx:discord/blah/watt-activation/part-202
  5. [5]Part 2114 facts
    ctx:discord/blah/watt-activation/part-211
  6. [6]Part 2036 facts
    ctx:discord/blah/watt-activation/part-203
  7. [7]Part 2289 facts
    ctx:discord/blah/watt-activation/part-228
  8. [8]Part 2164 facts
    ctx:discord/blah/watt-activation/part-216
  9. [9]Part 3318 facts
    ctx:discord/blah/watt-activation/part-331
  10. [10]2027 facts
    ctx:discord/blah/watt-activation/202
    • full textwatt-activation-202
      text/plain3 KBdoc:agent/watt-activation-202/bfa1acbb-7492-43ef-befc-99b62acad42a
      Show excerpt
      [2026-03-10 06:54] xenonfun: ``` lohe_spherical+lohe_v2 summary: train loss (last 50): 7.8510 ppl=2568.3 val loss (final): 7.8019 ppl=2445.1 sys_r @1000: 0.8817 global_r @1000: 0.2008 attn_r @1000:
  11. [11]2147 facts
    ctx:discord/blah/watt-activation/214
    • full textwatt-activation-214
      text/plain2 KBdoc:agent/watt-activation-214/6ebedb80-7ad3-4d5e-8432-4c87da7e9ad5
      Show excerpt
      [2026-03-11 04:17] xenonfun: ``` Model: lohe_spherical+lohe_v3 768d 12L 12H vocab=100277 Steps: 1000 BS=4 SEQ=1024 (4.1M tokens/run) Phase metrics sidecar → logs/phase_metrics_lohe_spherical_20260311_001319.jsonl ====================

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.