Revision History
Instruction Tuning · 12 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 claims that InstructGPT reduced bias, that Llama 3 post-training used PPO, an unsupported '10M examples / 25 categories' Llama 3 SFT figure, an overstated LIMA result ('near-GPT-4'), and a wrong description of Self-Instruct's classification step (plus 10 minor fixes)
Version 12 · Nov 1, 2026, 11:34 AM
- Improve internal links (2026-10-13)
Version 11 · Oct 13, 2026, 10:07 AM
- Added 6 contextual internal links
Version 10 · Jul 23, 2026, 03:04 PM
- Pre-update snapshot before enhancement to 4208 words.
Version 5 · May 18, 2026, 12:14 AM