Screenshot 2026-03-21 11.25.54 PM
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Screenshot 2026-03-21 11.25.54 PM has 35 facts recorded in Dontopedia across 20 references, with 5 live disagreements.
Mostly:rdf:type(18), has filename(4), filename(3)
Maturity scale
raw canonical shape-checked rule-derived certifiedRdf:typein disputerdf:type
- File[1]all time · 27
- File[2]all time · 7
- File[3]all time · 16
- Screenshot File[4]all time · 38
- Image File[5]all time · 37
- Screenshot File[6]all time · 54
- Image File[7]all time · 59
- Image File[8]all time · 67
- Image File[9]all time · 70
- Image File[10]all time · 6
Inbound mentions (15)
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.
referencesFileReferences File(3)
- Log Entry 2026 02 17 19 01
ex:log-entry-2026-02-17-19-01 - Log Entry 2026 03 11 22 38
ex:log-entry-2026-03-11-22-38 - Log Entry 9
ex:log-entry-9
hasAttachmentHas Attachment(2)
- Log Entry 4
ex:log-entry-4 - Message 3
ex:message-3
attachesFileAttaches File(1)
- Message 6
ex:message-6
containsAttachmentContains Attachment(1)
- Message 5
ex:message-5
containsReferenceToContains Reference to(1)
- Log Entry 2
ex:log-entry-2
mentionsFileMentions File(1)
- Log Entry 2026 04 22 15 55
ex:log-entry-2026-04-22-15-55
referencesResourceReferences Resource(1)
- Message 2026 01 16 20 23 01
ex:message-2026-01-16-20-23-01
sharedFileShared File(1)
- Xenonfun
ex:xenonfun
Other facts (10)
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 |
|---|---|---|
| Has Filename | Screenshot_2026-01-19_at_3.48.01_PM.png | [2] |
| Has Filename | Screenshot_2026-02-14_at_12.26.16_AM.png | [8] |
| Has Filename | Screenshot_2026-03-07_at_1.35.10_PM.png | [13] |
| Has Filename | Screenshot_2026-03-21_at_11.25.54_PM.png | [18] |
| Filename | Screenshot_2026-01-29_at_5.54.48_PM.png | [5] |
| Filename | Screenshot_2025-12-05_at-3.33.32_PM.png | [10] |
| Filename | Screenshot_2026-03-21_at_7.45.00_PM.png | [17] |
| File Name | Screenshot_2026-02-06_at_6.49.51_PM.png | [6] |
| File Name | Screenshot_2026-03-10_at_11.57.13_PM.png | [14] |
| Referenced in Message | Message 2026 01 29 22 55 | [5] |
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 (20)
ctx:discord/blah/random/27- full textrandom-27text/plain2 KB
doc:agent/random-27/e650c997-da27-4878-ba9f-a405e95b956aShow excerpt
[2026-02-17 18:23] xenonfun: yeah is with bpe, 7.5M model, with ~40MB of data on that (Gutenburg free library) I am going to do full training that should be enouge sample data now: ``` It's running! 55.7M tokens — so 1 epoch = 50.1M / 4096…
ctx:discord/blah/safiersemantics/7- full textsafiersemantics-7text/plain2 KB
doc:agent/safiersemantics-7/58260437-e784-4c66-b60f-0a0386674fd8Show excerpt
[2026-01-19 05:04] xenonfun: ``` [00:01:24 ERR] Failed to render tool HTML System.InvalidOperationException: View 'Pages/Shared/Components/_ToolCall' not found. Searched locations: at SafierSemantics.Services.ViewRenderService.RenderPa…
ctx:discord/blah/safiersemantics/16- full textsafiersemantics-16text/plain3 KB
doc:agent/safiersemantics-16/58f5296a-e209-4cc3-8461-a2eeca16a86fShow excerpt
[2026-01-20 20:40] foxhop.: That's a good sign typically. [2026-01-20 20:40] foxhop.: : ) [2026-01-20 20:41] xenonfun: yeah, think its fully ported to agent framework, has way better cleaner logging, and all the stuff from before works I th…
ctx:discord/blah/safiersemantics/38ctx:discord/blah/safiersemantics/37- full textsafiersemantics-37text/plain3 KB
doc:agent/safiersemantics-37/1bf7ce90-cb99-448a-95a1-806c99190cc2Show excerpt
[2026-01-29 15:04] xenonfun: Okay `master` branch is fully functional, resume even works correct https://youtu.be/uCT3By5_Fn4 next up is optimizing tool call output to be less verbose [2026-01-29 15:36] xenonfun: https://github.com/rsafie…
ctx:discord/blah/safiersemantics/54- full textsafiersemantics-54text/plain3 KB
doc:agent/safiersemantics-54/3405a9f6-99bd-44dd-ae7c-d7e08965d81bShow excerpt
[2026-02-06 23:10] xenonfun: it also is suppose to alrady have a valid AG-UI endpoint so could run your own UI off that, tho really not sure how well that works at this point with all the other stucture there [2026-02-06 23:10] traves_thebe…
ctx:discord/blah/safiersemantics/59- full textsafiersemantics-59text/plain3 KB
doc:agent/safiersemantics-59/66ed50cb-adf4-467b-bcd7-58dab42c65d1Show excerpt
[2026-02-08 16:50] xenonfun: this I think has unfucked itself, but will have to look at diff, it was already messy not convinced this was improvement. will probablt go look at this simpled down template apps [2026-02-08 17:01] xenonfun: (f…
ctx:discord/blah/safiersemantics/67- full textsafiersemantics-67text/plain3 KB
doc:agent/safiersemantics-67/88338f87-110b-46da-a4ad-fe0eadef124aShow excerpt
[2026-02-13 15:44] xenonfun: ``` The user wants me to: Take the layered-accounting-logo-primary.png logo design Create a Three.js version of it with 3 rotating elements Middle element rotates counter to top and bottom Create a sample page …
ctx:discord/blah/safiersemantics/70- full textsafiersemantics-70text/plain3 KB
doc:agent/safiersemantics-70/dbacde78-f635-4864-93c8-c2425e32c560Show excerpt
[2026-02-19 20:25] xenonfun: model-ds being trained, asked it to optimize just on this training set what can be done without blowing out my 24GB limit and not exhausting the model from not enough data. (files: Screenshot_2026-02-19_at_3.23.…
ctx:discord/blah/tpmjs/6- full texttpmjs-6text/plain3 KB
doc:agent/tpmjs-6/bfaa5404-dabc-4f9a-b804-a4b69c39ae77Show excerpt
[2025-12-04 16:42] ajaxdavis: - they get a playground to test their tools (freeish) - searchAndUseTpmjsTool (might be useful) but nothing really, just chicken and egg problem [2025-12-04 16:42] ajaxdavis: in a race against vercel regist…
ctx:discord/blah/tpmjs/20- full texttpmjs-20text/plain2 KB
doc:agent/tpmjs-20/87520c47-6b07-4864-8de7-140bd70e926bShow excerpt
[2026-01-14 06:32] ajaxdavis: (the still butchering the tpmjs stuff, trying to figure out the ownership model of collections/agents, i've currently settled on that you "fork" them. or "copy") it's currently a clone model and it becomes your…
ctx:discord/blah/tpmjs/28- full texttpmjs-28text/plain3 KB
doc:agent/tpmjs-28/acb1d01e-93d9-458c-9585-0143026f36deShow excerpt
[2026-01-16 20:18] xenonfun: from the tools response (files: Screenshot_2026-01-16_at_3.18.33_PM.png) [2026-01-16 20:18] xenonfun: I will check working one [2026-01-16 20:19] ajaxdavis: also the health check tool on unsandbox probs doesnt w…
ctx:discord/blah/watt-activation/76- full textwatt-activation-76text/plain3 KB
doc:agent/watt-activation-76/961fc69a-4972-401d-be24-5f9157949bafShow excerpt
[2026-03-07 18:31] xenonfun: ``` Excellent results. Full-sequence chunking (chunk_size=L) with the new formulation: ┌────────────────┬──────────────┬─────────────────────┬─────────────┬───────────┬────────┐ │ Config │ Old 5D to…
ctx:discord/blah/watt-activation/209- full textwatt-activation-209text/plain2 KB
doc:agent/watt-activation-209/d7eabcf9-d506-432f-9a2c-b25cf5ef8ccdShow excerpt
[2026-03-11 03:51] xenonfun: ```# coupling (K) and adjacency are structural constants — not updated. 598 + # v2: update harmonic_coeffs (G×H learned frequency weights) 599 + # v3: update mode_amplitudes (k scalars,…
ctx:discord/blah/watt-activation/274- full textwatt-activation-274text/plain1 KB
doc:agent/watt-activation-274/2c5f3f8e-4dd2-4aea-8be5-e9d163af5028Show excerpt
[2026-03-13 20:06] xenonfun: It tried its fuckery again but we learned an antipattern to remember. ```⏺ Here's the difference: _normalize (attention.py:136-141): norm = mx.sqrt(mx.maximum(sum(x²), eps)) # eps=1e-6 for float32 x / no…
ctx:discord/blah/watt-activation/340- full textwatt-activation-340text/plain3 KB
doc:agent/watt-activation-340/ec99b9a7-182e-4e78-864e-12381357aa47Show excerpt
[2026-03-15 21:48] xenonfun: (files: Screenshot_2026-03-15_at_5.48.31_PM.png) [2026-03-15 21:58] xenonfun: try some weird stuff (files: Screenshot_2026-03-15_at_5.58.32_PM.png) [2026-03-15 22:32] xenonfun: We measure bandwidth now, how muc…
ctx:discord/blah/watt-activation/482- full textwatt-activation-482text/plain2 KB
doc:agent/watt-activation-482/3ab5810b-8f32-47b9-a3a9-debf4ee57bd3Show excerpt
[2026-03-21 23:22] xenonfun: ``` ⏺ The analytical path reads raw bytes — it never uses the NPY loader. That means the BPE-8K experiment actually trained on the raw bytes of the NPY file (header + int32 data), not on the tokens themselves.…
ctx:discord/blah/watt-activation/486- full textwatt-activation-486text/plain3 KB
doc:agent/watt-activation-486/c8568fef-e9f2-4d48-9840-89f375514ea3Show excerpt
[2026-03-22 03:19] xenonfun: ``` ⏺ The IVF was useless before because ivf_dirty got set on every insert (128/step), and only cleared every 500 steps on rebuild. So 99.7% of lookups fell back to linear scan. Now: - New entries get assigne…
ctx:discord/blah/watt-activation/682- full textwatt-activation-682text/plain3 KB
doc:agent/watt-activation-682/d73a60c6-dcef-4901-b083-537925f7bdfeShow excerpt
[2026-04-22 15:51] xenonfun: (files: Screenshot_2026-04-22_at_11.51.23_AM.png) [2026-04-22 15:54] xenonfun: done 20000 steps in 7006.5s — best bpb=1.3575 -> data/runs/unified_rotor_gpu_25m/final [2026-04-22 15:55] xenonfun: (files: Scree…
ctx:discord/blah/watt-activation/693- full textwatt-activation-693text/plain3 KB
doc:agent/watt-activation-693/8372b985-fcfe-4f68-a730-4cd9d14a0fb9Show excerpt
[2026-04-28 23:22] xenonfun: KickModel (sparse extreme — top pick for fast iteration) 25 params total at g=8 (yes, twenty-five). BPB 2.04 on fineweb. Architecture: ∂ₜΨ = [Ω, Ψ] − γ⟨Ψ⟩ + kick, G independent S³ oscillators (3G+1 params), exp…
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.