Dontopedia

AMX

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)

AMX has 20 facts recorded in Dontopedia across 9 references, with 2 live disagreements.

20 facts·15 predicates·9 sources·2 in dispute

Mostly:rdf:type(3), has execution mode(2), wins for(1)

Maturity scale raw canonical shape-checked rule-derived certified

Inbound mentions (11)

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.

abbreviatedAsAbbreviated As(1)

comparesCompares(1)

comparesToCompares to(1)

didntMultithreadDidnt Multithread(1)

hasTechnologyHas Technology(1)

isCpuBoundOnIs Cpu Bound on(1)

lacksFeatureLacks Feature(1)

lacksMultithreadingLacks Multithreading(1)

losesToLoses to(1)

overOver(1)

statusOfStatus of(1)

Other facts (18)

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.

18 facts
PredicateValueRef
Rdf:typeHardware Feature[7]
Rdf:typeTechnology[8]
Rdf:typeHardware Feature[9]
Has Execution Modesynchronous[8]
Has Execution Modein-process[8]
Wins forSmall Matrices[1]
Is Cpu Bound25m Model[1]
Is BaselineApple Matrix Extension[2]
Is Superior to MetalMetal[3]
Has Zero Dispatchtrue[3]
Achieves Gflop S500+[3]
Is Synchronoustrue[3]
Multithreading Statusnot yet[4]
Needs Multithreadingtrue[4]
Historical Tech ContextIntel Amx[5]
Essential for Cpu Speedtrue[6]
Has Dispatch Overheadzero[8]
Has Performance Attributeextremely fast at sgemm at d=560[9]

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.

winsForblah/watt-activation/part-638
ex:small-matrices
isCpuBoundblah/watt-activation/part-638
ex:25m-model
isBaselineblah/watt-activation/part-634
ex:apple-matrix-extension
isSuperiorToMetalblah/watt-activation/part-641
ex:metal
hasZeroDispatchblah/watt-activation/part-641
true
achievesGflopSblah/watt-activation/part-641
500+
isSynchronousblah/watt-activation/part-641
true
multithreadingStatusblah/watt-activation/part-640
not yet
needsMultithreadingblah/watt-activation/part-640
true
historicalTechContextblah/watt-activation/part-665
ex:intel-amx
essentialForCpuSpeedblah/watt-activation/part-678
true
typeblah/watt-activation/637
ex:HardwareFeature
typeblah/watt-activation/638
ex:Technology
labelblah/watt-activation/638
AMX
hasDispatchOverheadblah/watt-activation/638
zero
hasExecutionModeblah/watt-activation/638
synchronous
hasExecutionModeblah/watt-activation/638
in-process
typeblah/watt-activation/675
ex:HardwareFeature
labelblah/watt-activation/675
AMX
hasPerformanceAttributeblah/watt-activation/675
extremely fast at sgemm at d=560

References (9)

9 references
  1. [1]Part 6382 facts
    ctx:discord/blah/watt-activation/part-638
  2. [2]Part 6341 fact
    ctx:discord/blah/watt-activation/part-634
  3. [3]Part 6414 facts
    ctx:discord/blah/watt-activation/part-641
  4. [4]Part 6402 facts
    ctx:discord/blah/watt-activation/part-640
  5. [5]Part 6651 fact
    ctx:discord/blah/watt-activation/part-665
  6. [6]Part 6781 fact
    ctx:discord/blah/watt-activation/part-678
  7. [7]6371 fact
    ctx:discord/blah/watt-activation/637
    • full textwatt-activation-637
      text/plain3 KBdoc:agent/watt-activation-637/19b1b41c-2a51-4202-8887-489a5512853c
      Show excerpt
      [2026-04-16 00:35] xenonfun: Proposed new order 1. Batching fix for training path (quick) — apply the same batched_linear_forward change to forward_train and its backward. Free 2.5× on training throughput. This must happen before Optio
  8. [8]6385 facts
    ctx:discord/blah/watt-activation/638
    • full textwatt-activation-638
      text/plain3 KBdoc:agent/watt-activation-638/539b1483-cc08-42f3-9ea1-6a7f6ae47ab8
      Show 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
  9. [9]6753 facts
    ctx:discord/blah/watt-activation/675
    • full textwatt-activation-675
      text/plain2 KBdoc:agent/watt-activation-675/328d1b65-525d-44a4-8d22-56a80354a618
      Show excerpt
      [2026-04-21 23:43] xenonfun: hmm well that didn't work well: ``` ⏺ Honest smoketest result — not the number I was hoping to see: ┌──────────────────────┬────────┬───────┬────────┬────────────────┐ │ Path │ BPB │ Time

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.