throughput gap
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-07-04.)
throughput gap is G=4, H=4 small model.
Mostly:caused by(2), is0 22x(1), caused by python loops(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (1)
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.
summarizesThroughputGapRootCauseSummarizes Throughput Gap Root Cause(1)
- Xenonfun Message 2026 03 10 06 55
ex:xenonfun-message-2026-03-10-06-55
Other facts (16)
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 |
|---|---|---|
| Caused by | Kv Inp 4x Larger | [2] |
| Caused by | Python Head Loop | [3] |
| Is0 22x | 0.22x | [1] |
| Caused by Python Loops | true | [1] |
| Expected for Short Sequences | true | [1] |
| Is Expected | true | [1] |
| Is Priority to Address Next | null | [2] |
| Attributed to Implementation | null | [3] |
| Rdf:type | Performance Metric | [4] |
| Has Value | 0.22 | [4] |
| Has Context | PyTorch benchmark | [4] |
| Has Cause | short sequences with Python-level loops | [4] |
| Has Percentage Value | 12 | [5] |
| Attributed to | Python Head Loop | [5] |
| Description | G=4, H=4 small model | [6] |
| Future Behavior | at GPT-2 scale the gap will be smaller | [6] |
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 (6)
ctx:discord/blah/watt-activation/part-48ctx:discord/blah/watt-activation/part-203ctx:discord/blah/watt-activation/part-451ctx:discord/blah/watt-activation/48- full textwatt-activation-48text/plain3 KB
doc:agent/watt-activation-48/17fbe6b3-28dd-44cc-9e32-ece98c3e9a74Show excerpt
[2026-03-07 06:49] xenonfun: ``` All 6 configs completed successfully — no crashes. Full summary: Complex Oscillator Sweep Results (20K iters, 8K vocab) ┌──────────────────┬───────────┬──────────┬────────────────────────────────┐ │ …
ctx:discord/blah/watt-activation/449- full textwatt-activation-449text/plain2 KB
doc:agent/watt-activation-449/f6da4410-b0fa-46d2-9cb5-99d205dd2e55Show excerpt
[2026-03-21 03:48] xenonfun: ⏺ You're right. At step 950: ``` ┌──────────┬───────────┬──────────┐ │ Metric │ Helmholtz │ v3 │ ├──────────┼───────────┼──────────┤ │ BPB │ 2.257 │ 2.257 │ ├──────────┼───────────┼…
ctx:claims/document/030a0648-513f-45a0-a9a4-618fb7d44634- full textxenonfun: Smoke test results (d=256, 4L, G=4, H=4): - r_global at random init: 0.69 (wittext/plain675 B
discord:convmsg/5853Show excerpt
xenonfun: Smoke test results (d=256, 4L, G=4, H=4): - r_global at random init: 0.69 (within-block sync already present) - Mean r_global during training: 0.61 - Loss descent: 0.176 at 100 steps — matches spectral+lohe_v2 closely - Th…
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.