Revision History
Batch · 10 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the typical language-model batch size, which was given as 256 to 8,192 tokens when LLM pretraining batches run to millions of tokens (3.2M for GPT-3 175B) (plus 10 minor fixes)
Version 10 · Nov 5, 2026, 11:59 AM
- Improve internal links (2026-10-13)
Version 9 · Oct 13, 2026, 10:01 AM
- Added 6 contextual internal links
Version 8 · Jul 23, 2026, 03:00 PM
- Render formulas with LaTeX math notation
Version 7 · Jul 11, 2026, 10:31 PM