128K run
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
128K run has 21 facts recorded in Dontopedia across 2 references.
Mostly:has fewer optimizer steps than(2), presupposes underperformance(1), warmup as fraction of epoch(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (15)
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.
isRecommendedFixForIs Recommended Fix for(4)
- Fix Lower Base Lr
ex:fix-lower-base-lr - Fix Lower Rotational Strength
ex:fix-lower-rotational-strength - Fix Multi Epoch Schedule
ex:fix-multi-epoch-schedule - Fix Shorter Warmup
ex:fix-shorter-warmup
providesBetterGradientDiversityThanProvides Better Gradient Diversity Than(2)
causesPoorPerformanceInCauses Poor Performance in(1)
- Warmup Schedule Mismatch
ex:warmup-schedule-mismatch
hasMoreLocalContextsPerStepThanHas More Local Contexts Per Step Than(1)
- 2k Run
ex:2k-run
potentiallyInappropriateForPotentially Inappropriate for(1)
- Decay Settings
ex:decay-settings
providesMoreIterativeRefinementThanProvides More Iterative Refinement Than(1)
- 2k Run
ex:2k-run
reducesIterativeRefinementInReduces Iterative Refinement in(1)
- Fewer Optimizer Steps
ex:fewer-optimizer-steps
statedConditionalActionStated Conditional Action(1)
- Xenonfun
ex:xenonfun
unsuitableForUnsuitable for(1)
- Hyperparameters
ex:hyperparameters
Other facts (20)
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 |
|---|---|---|
| Has Fewer Optimizer Steps Than | 2k Run | [1] |
| Has Fewer Optimizer Steps Than | 2k Run | [2] |
| Presupposes Underperformance | Relative to 2k Run | [1] |
| Warmup As Fraction of Epoch | about 32% | [1] |
| Uses Sequence Length | 131072 | [1] |
| Uses Batch Size | 1 | [1] |
| Has Steps Per Epoch | 1554 | [1] |
| Underperforms Compared to | 2k Run | [1] |
| Involves One Giant Window Per Update | True | [1] |
| Spent Massive Fraction Under Ramped | Training | [1] |
| Steps Per Epoch | 1554 | [2] |
| Warmup Ratio Approximate | 0.32 | [2] |
| Trained Under Ramped | true | [2] |
| Context Length | 131072 | [2] |
| Pays Optimization Cost Without Proportionate Signal | true | [2] |
| Has Worse Batch Diversity Than | 2k Run | [2] |
| Batch Size | 1 | [2] |
| Sequence Length | 131072 | [2] |
| Has One Giant Window Per Update | true | [2] |
| Position Handling Difficulty | high | [2] |
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 (2)
ctx:discord/blah/watt-activation/part-265ctx:discord/blah/watt-activation/263- full textwatt-activation-263text/plain2 KB
doc:agent/watt-activation-263/18f755d3-7fe3-4c11-aca5-c67fdc4eb174Show excerpt
[2026-03-13 04:59] xenonfun: ``` • Yes. Several plausible reasons, even without blaming the architecture. Most important ones: - Warmup/schedule mismatch - the 2K run had 12,432 steps/epoch - the 128K run had only 1,554 …
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.