d_model
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
d_model has 14 facts recorded in Dontopedia across 8 references, with 1 live disagreement.
Mostly:rdf:type(4), must be greater or equal to(1), provides headroom for context mixing(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (6)
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.
staysAtDimensionStays at Dimension(2)
- Output Proj
ex:output-proj - V
ex:v
evaluatesEfficiencyEvaluates Efficiency(1)
- Scaling Observation
ex:scaling-observation
hasNoDModelHas No D Model(1)
- Resonantwirelm
ex:resonantwirelm
lacksDmodelLacks Dmodel(1)
- Resonantwirelm
ex:resonantwirelm
toDimTo Dim(1)
- Linear Layer
ex:linear-layer
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 |
|---|---|---|
| Rdf:type | Model Dimension | [5] |
| Rdf:type | Hyperparameter | [6] |
| Rdf:type | Hyperparameter | [7] |
| Rdf:type | Hyperparameter | [8] |
| Must Be Greater or Equal to | G×H × 4-8 | [1] |
| Provides Headroom for Context Mixing | true | [1] |
| Example Value | 832 | [2] |
| Is Projection Target | Current Pipeline | [3] |
| Is Least Efficient to Scale | Scaling Lever | [4] |
| Scaling Helps | true | [7] |
| Scaling Effect | fill the GPU | [7] |
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 (8)
ctx:discord/blah/watt-activation/part-323ctx:discord/blah/watt-activation/part-322ctx:discord/blah/watt-activation/part-321ctx:discord/blah/watt-activation/part-355ctx:discord/blah/watt-activation/319- full textwatt-activation-319text/plain2 KB
doc:agent/watt-activation-319/f54ddf34-a21b-47fb-8296-277054f2ccaaShow excerpt
[2026-03-15 02:58] lisamegawatts: You're right — the whole point of QPSK isn't to be sparse, it's to be bandwidth-efficient. In telecom, QPSK packs 2 bits into one symbol period because the receiver only needs to distinguish 4 phase states,…
ctx:discord/blah/watt-activation/353- full textwatt-activation-353text/plain3 KB
doc:agent/watt-activation-353/cc7a24c1-66ae-472e-a74c-30bb70fe2a69Show excerpt
[2026-03-17 09:19] xenonfun: ``` ============================================================ K4_cur10 K=4 curriculum=10% ============================================================ step 1000/5000 BPB=3.173 719,581 tok/s step 2…
ctx:discord/blah/watt-activation/355- full textwatt-activation-355text/plain3 KB
doc:agent/watt-activation-355/e62c81a8-1082-4c07-b675-3759a8600d0eShow excerpt
[2026-03-17 15:27] xenonfun: ``` Key findings: 1. Depth scaling is smooth and strong: BPB drops monotonically 3.00→2.53 from D=6→D=32. DC@16 rises 72%→91%. 2. Retrieval reach = 128 for ALL configs — every model retrieves across the f…
ctx:discord/blah/watt-activation/392- full textwatt-activation-392text/plain1 KB
doc:agent/watt-activation-392/ddbc1ecc-c3dc-4acc-b74a-05cca6e39186Show excerpt
[2026-03-19 03:47] xenonfun: ``` ❯ do we not have concept of SNR and bandwidth in these or they pure rotation and geometry? ⏺ They're pure rotation and geometry right now — no SNR or bandwidth concepts. The ResonantWireLM has no attention,…
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.