Evaluation 1
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Evaluation 1 has 25 facts recorded in Dontopedia across 3 references, with 3 live disagreements.
Mostly:compares to(2), possible cause of gap(2), requires reproduction of(2)
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.
evaluatedPerformanceEvaluated Performance(1)
- Xenonfun
ex:xenonfun
Other facts (25)
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 |
|---|---|---|
| Compares to | Gpt 2 Training | [1] |
| Compares to | Run Config 5 | [2] |
| Possible Cause of Gap | forward logit parity compounding | [3] |
| Possible Cause of Gap | dataset version drift | [3] |
| Requires Reproduction of | per-batch averaging | [3] |
| Requires Reproduction of | bf16/tf32 quirks | [3] |
| Assesses Duration | 13.5 days | [1] |
| Hardware Config | single consumer GPU | [1] |
| Parameter Count | 145000000 | [1] |
| Training Token Count | 9000000000 | [1] |
| Opinion | not bad at all | [1] |
| Entity Evaluated | Learned Gate | [2] |
| Plateau Value | 4.58 | [2] |
| Worse Than | Run Config 5 | [2] |
| Reported in | Xenonfun | [3] |
| Has Metric | 1.4192 | [3] |
| Metric Name | eval BPB | [3] |
| Metric Condition | python-stride | [3] |
| Gap to Reference | 0.0041 | [3] |
| Previous Gap | 0.0072 | [3] |
| Compounding Over Tokens | 114000 | [3] |
| Reference Timestamp Context | 1.4151 was logged | [3] |
| Requires Reproduction for Zero Gap | true | [3] |
| Reproduction Is Worth Chasing | false | [3] |
| Parity Band | ~0.01 | [3] |
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 (3)
ctx:discord/blah/watt-activation/94- full textwatt-activation-94text/plain3 KB
doc:agent/watt-activation-94/3ac27cdb-b74d-48fa-9f59-8e2eb7c9c9adShow excerpt
[2026-03-08 02:30] xenonfun: okay so quick estimate how long would it take for us to train a full 9B tokens if we had them? I'm getting something like 13.5 days at 7.7Ktok/s ? ⏺ Your math checks out. 9B tokens / 7.7K tok/s = 1,168,831 seco…
ctx:discord/blah/watt-activation/409- full textwatt-activation-409text/plain2 KB
doc:agent/watt-activation-409/27cf5bf6-23d3-4bb2-8da9-9ee785b7e6a7Show excerpt
[2026-03-19 13:28] xenonfun: ``` ⏺ Interesting contrast: ┌─────┬──────┬──────┬──────┬─────────┬──────┐ │ LR │ Step │ BPB │ r │ SNR │ DC │ ├─────┼──────┼──────┼──────┼─────────┼──────┤ │ 0.1 │ 350 │ 4.58 │ 0.41 │ -6.9dB …
ctx:discord/blah/watt-activation/633- full textwatt-activation-633text/plain2 KB
doc:agent/watt-activation-633/98cb2cdc-0a8e-4624-a978-d2d562057e94Show excerpt
[2026-04-15 23:58] lisamegawatts: yours is way faster to train, the data sets are all in hf now, should do a 50m chinchilla optimal, i think we have enough data, use cleaned corpus and the synth ones but skip text pipeline probably as it di…
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.