Iteration 100
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Iteration 100 has 31 facts recorded in Dontopedia across 4 references, with 2 live disagreements.
Mostly:iteration number(3), has learning rate(2), loss value(2)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (3)
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.
consistsOfConsists of(1)
- Training Run Linear Seq2048
ex:training-run-linear-seq2048
hasHigherLossThanHas Higher Loss Than(1)
- Iteration 10
ex:iteration-10
temporallyPrecedesTemporally Precedes(1)
- Iteration 90
ex:iteration-90
Other facts (31)
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 |
|---|---|---|
| Iteration Number | 100 | [2] |
| Iteration Number | 100 | [3] |
| Iteration Number | 100 | [4] |
| Has Learning Rate | 0.00003 | [1] |
| Has Learning Rate | 0.00003 | [4] |
| Loss Value | nan | [2] |
| Loss Value | 3.1939 | [3] |
| Learning Rate | 0.00003 | [2] |
| Learning Rate | 0.00003 | [3] |
| Rdf:type | Iteration | [3] |
| Rdf:type | Iteration | [4] |
| Belongs to | Training Run Linear Seq2048 | [1] |
| Has Gpu Cache Memory | 34.8GB | [1] |
| Has Gpu Memory | 1.0GB | [1] |
| Has Gpu Peak Memory | 34.3GB | [1] |
| Has Iteration Number | 100 | [1] |
| Has Loss | 3.1939 | [1] |
| Has Process Memory | 3.2GB | [1] |
| Has Tokens Per Second | 6763 | [1] |
| Tokens Per Second | 6763 | [3] |
| Gpu Memory Usage Gb | 1 | [3] |
| Gpu Memory Peak Gb | 34.3 | [3] |
| Gpu Cache Gb | 34.8 | [3] |
| Process Memory Gb | 3.2 | [3] |
| Part of Run | Training Run Grouped1 | [4] |
| Has Loss Value | 3.1608 | [4] |
| Has Token Throughput | 17618 | [4] |
| Gpu Memory Usage | 1 | [4] |
| Gpu Memory Peak | 12 | [4] |
| Gpu Cache Memory | 11.8 | [4] |
| Process Memory Usage | 3.2 | [4] |
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 (4)
ctx:discord/blah/watt-activation/part-114ctx:discord/blah/watt-activation/111ctx:discord/blah/watt-activation/114- full textwatt-activation-114text/plain2 KB
doc:agent/watt-activation-114/bc8b5ab5-bbc4-4aea-ac89-9211a96127c3Show excerpt
[2026-03-08 22:42] xenonfun: Now add val evaluation in the training loop, after the checkpoint save block [2026-03-08 22:44] xenonfun: ``` === linear (seq=2048) === Training HarmonicGPT | attn=linear d=768 L=12 H=12 seq=2048 Parameters: 86,…
ctx:discord/blah/watt-activation/115- full textwatt-activation-115text/plain3 KB
doc:agent/watt-activation-115/105443fa-23c3-42fc-a3a6-065e0c10a484Show excerpt
[2026-03-08 22:44] xenonfun: ``` === grouped1 (seq=2048) === Training HarmonicGPT | attn=grouped1 d=768 L=12 H=12 seq=2048 Parameters: 86,825,760 Loaded 5,000 train sequences of length 2048 iter 10/100 | loss 5.4746 | lr 3.00e-04 | tok/…
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.