2K run
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
2K run has 21 facts recorded in Dontopedia across 4 references.
Mostly:has same token budget as(2), provides better gradient diversity than(2), serves as baseline(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (10)
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.
appropriateForAppropriate for(1)
- Decay Settings
ex:decay-settings
commitsToTestingCommits to Testing(1)
- Xenonfun
ex:xenonfun
hasWorseBatchDiversityThanHas Worse Batch Diversity Than(1)
- 128k Run
ex:128k-run
plansToRunPlans to Run(1)
- Xenonfun
ex:xenonfun
resumedFrom2kResumed From2k(1)
- 10k Run
ex:10k-run
tunedForTuned for(1)
- Hyperparameters
ex:hyperparameters
tunedOnTuned on(1)
- Hyperparameters
ex:hyperparameters
underperformsComparedToUnderperforms Compared to(1)
- 128k Run
ex:128k-run
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 Same Token Budget As | 128k Run | [1] |
| Has Same Token Budget As | 128k Run | [3] |
| Provides Better Gradient Diversity Than | 128k Run | [1] |
| Provides Better Gradient Diversity Than | 128k Run | [3] |
| Serves As Baseline | For Comparison | [1] |
| Warmup As Fraction of Epoch | about 4% | [1] |
| Uses Sequence Length | 2048 | [1] |
| Has Steps Per Epoch | 12432 | [1] |
| Provides Many More Local Contexts Per Step | True | [1] |
| Uses Batch Size | 8 | [1] |
| Tests Fixes | Coupling Fixes | [2] |
| Likely Long Running | Long Computation | [2] |
| Uses Optimizations | Coupling Optimizations | [2] |
| Steps Per Epoch | 12432 | [3] |
| Warmup Ratio Approximate | 0.04 | [3] |
| Provides More Iterative Refinement Than | 128k Run | [3] |
| Batch Size | 8 | [3] |
| Sequence Length | 2048 | [3] |
| Has More Local Contexts Per Step Than | 128k Run | [3] |
| Rdf:type | Execution Run | [4] |
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 (4)
ctx:discord/blah/watt-activation/part-265ctx:discord/blah/watt-activation/part-446ctx: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 …
ctx:discord/blah/watt-activation/444- full textwatt-activation-444text/plain2 KB
doc:agent/watt-activation-444/885ad207-9977-4a41-b00d-589966869135Show excerpt
[2026-03-20 17:49] xenonfun: <@1211062099137265723>, had it reread your FedSym merge, think this is closer to what you desribed and has zero scaling issues. ``` ⏺ Now I see it. The key things I missed: 1. _transfer_weights_parallel with …
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.