L1024
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
L1024 has 12 facts recorded in Dontopedia across 4 references.
Mostly:estimated iterations per second(1), gate matrix memory forward(1), estimated time for20k iters(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (8)
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.
hasMemoryUsageForSeqLenHas Memory Usage for Seq Len(1)
- Anchor V3
ex:anchor-v3
hasSequenceLengthHas Sequence Length(1)
- L1024 Scaling Run
ex:l1024-scaling-run
reportsDataForReports Data for(1)
- Xenonfun
ex:xenonfun
runsSequenceLengthRuns Sequence Length(1)
- Bp3qejkfr
ex:bp3qejkfr
runsSequenceLengthsRuns Sequence Lengths(1)
- Bench Seqlen Scaling Py
ex:bench-seqlen-scaling-py
scalesFromScales From(1)
- Akan M32 Scaling Results
ex:akan-m32-scaling-results
scalesFromSeqLenScales From Seq Len(1)
- Akan M32 Scaling
ex:akan-m32-scaling
showsSaturationAtL1024PlusShows Saturation at L1024 Plus(1)
- Scaling Benchmark
ex:scaling-benchmark
Other facts (11)
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 |
|---|---|---|
| Estimated Iterations Per Second | ~40-55 | [1] |
| Gate Matrix Memory Forward | 64MB (4 layers × [1,1024,1024,4]) | [1] |
| Estimated Time For20k Iters | 6-8 min | [1] |
| Seq Len Value | 1024 | [1] |
| Has Peak Memory | 1.7GB | [2] |
| Has Tokens Per Second | 80.7K | [2] |
| Has Iterations Per Second | 79 | [2] |
| Has Active Memory | 139MB | [2] |
| Has Compile Time | — | [2] |
| Has Drift | 0 | [2] |
| Succeeds to | L2048 | [3] |
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-57ctx:discord/blah/watt-activation/part-81ctx:discord/blah/watt-activation/part-80ctx:discord/blah/watt-activation/58- full textwatt-activation-58text/plain3 KB
doc:agent/watt-activation-58/2b893cf3-b6f9-4fdb-a94d-202a521b9459Show excerpt
[2026-03-07 09:29] xenonfun: ``` Anchor v3 running. On memory — the anchor overhead is minimal: Anchor v3 extra parameters per layer (m=32): - anchor_re/im: 32 × 4 phases = 256 floats - anchor_omega: 256 floats - W_anchor_q: 320 ×…
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.