Dontopedia

Final Training Stats

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-05-02.)

Final Training Stats has 11 facts recorded in Dontopedia across 1 reference.

11 facts·11 predicates·1 sources

Mostly:belongs to(1), derived from(1), has average tokens per second(1)

Maturity scale raw canonical shape-checked rule-derived certified

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.

11 facts
PredicateValueRef
Belongs toTraining Run Linear Seq2048[1]
Derived FromIteration Metrics[1]
Has Average Tokens Per Second6750[1]
Has Elapsed Time Seconds121.4[1]
Has Final Loss3.1939[1]
Has Gpu Active Memory1.01 GB[1]
Has Gpu Cache Memory34.75 GB[1]
Has Gpu Peak Memory34.31 GB[1]
Has Process Rss Memory4.16 GB[1]
Has Total Tokens819200[1]
Shows Memory IncreaseProcess Memory[1]

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.

belongsToblah/watt-activation/part-114
ex:training-run-linear-seq2048
derivedFromblah/watt-activation/part-114
ex:iteration-metrics
hasAverageTokensPerSecondblah/watt-activation/part-114
6750
hasElapsedTimeSecondsblah/watt-activation/part-114
121.4
hasFinalLossblah/watt-activation/part-114
3.1939
hasGpuActiveMemoryblah/watt-activation/part-114
1.01 GB
hasGpuCacheMemoryblah/watt-activation/part-114
34.75 GB
hasGpuPeakMemoryblah/watt-activation/part-114
34.31 GB
hasProcessRssMemoryblah/watt-activation/part-114
4.16 GB
hasTotalTokensblah/watt-activation/part-114
819200
showsMemoryIncreaseblah/watt-activation/part-114
ex:process-memory

References (1)

1 references
  1. [1]Part 11411 facts
    ctx:discord/blah/watt-activation/part-114

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.