Revision History
Scaling Laws for Neural Language Models · 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: Fixed the wrong author attribution of the 2024 Kaplan-Chinchilla reconciliation paper (Pearce and Song, not Porian et al.), corrected the per-10x loss-reduction percentages (about 16%, 20% and 11%, not 5.3%, 6.4% and 3.4%), stated that all ten authors were on the GPT-3 paper, corrected the claim that Kaplan and Chinchilla share a loss formula, and corrected the description of LLaMA as Chinchilla-optimal (plus 10 minor fixes)
Version 10 · Nov 5, 2026, 05:03 AM
- Improve internal links (2026-10-13)
Version 9 · Oct 13, 2026, 10:10 AM
- Added 6 contextual internal links
Version 8 · Jul 23, 2026, 03:18 PM
- Render formulas with LaTeX math notation
Version 7 · Jul 12, 2026, 12:37 AM