Metal Subagent
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Metal Subagent has 5 facts recorded in Dontopedia across 2 references.
Mostly:dispatches one sgemm per(1), flags next lever(1), has prior report(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (2)
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.
addressedToAddressed to(1)
- Attention Sgemm Batching
ex:attention-sgemm-batching
referencesSubagentFlagReferences Subagent Flag(1)
- Xenonfun
ex:xenonfun
Other facts (5)
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 |
|---|---|---|
| Dispatches One Sgemm Per | Matmul Call | [1] |
| Flags Next Lever | Batch Attention Sgemms | [1] |
| Has Prior Report | Matmul Tiled | [1] |
| Rdf:type | Agent | [2] |
| Dispatched Operation | Sgemm | [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-641ctx:discord/blah/watt-activation/638- full textwatt-activation-638text/plain3 KB
doc:agent/watt-activation-638/539b1483-cc08-42f3-9ea1-6a7f6ae47ab8Show excerpt
[2026-04-16 00:45] xenonfun: ``` ⏺ Let me answer the Python question first, then look at rayon parallelism. Python training setup (from scaling_wave_native_curriculum.py): - BATCH_SIZE = 32 (base script has BATCH_SIZE=16 with GRAD_ACCU…
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.