56,966 blogs · [ { "id": "01a087e0-8b2d-700b-bcda-6f67f16d9917", "title": "Towards a Bitter Lesson of Optimization: When Neural Networks Write Their Own Update Rules", "url": "https://mosthumble.github.io/posts/Towards-a-Bitter-Lesson-of-Optimization-When-Neural-Networks-Write-Their-Own-Update-Rules/", "published_at": "2026-04-07T14:47:50+00:00" }, { "id": "01a087e0-8b2d-700b-bcda-6f67f1925ade", "title": "Why Modern LLMs Dropped Mean Centering (And Got Away With It)", "url": "https://mosthumble.github.io/posts/Why-Modern-LLMs-Dropped-Mean-Centering-(And-Got-Away-With-It)/", "published_at": "2026-02-22T01:00:00+00:00" }, { "id": "01a087e0-8b2d-700b-bcda-6f67f1c82cc1", "title": "The Epsilon Trap: When Adam Stops Being Adam", "url": "https://mosthumble.github.io/posts/The-Epsilon-Trap-When-Adam-Stops-Being-Adam/", "published_at": "2026-01-17T06:30:00+00:00" }, { "id": "01a087e0-8b2d-700b-bcda-6f67f2468d0b", "title": "Entropic Instruction Following: Does Semantic Coherence Help LLMs Follow Instructions?", "url": "https://mosthumble.github.io/posts/Entropic-Instruction-Following/", "published_at": "2025-12-02T08:01:00+00:00" }, { "id": "01a087e0-8b2d-700b-bcda-6f67f26b7f56", "title": "Elements Of Mechanistic Interpretability: From Observation to Causation", "url": "https://mosthumble.github.io/posts/Elements-Of-Mechanistic-Interpretability-From-Observation-to-Causation/", "published_at": "2025-10-26T15:45:00+00:00" } ] posts Claim your blog

Blog · corpus.blog/blogs/mosthumble.github.io/posts

mosthumble.github.io

2026

Towards a Bitter Lesson of Optimization: When Neural Networks Write Their Own Update Rules original ↗

7 Apr 2026

Why Modern LLMs Dropped Mean Centering (And Got Away With It)original ↗

22 Feb 2026

The Epsilon Trap: When Adam Stops Being Adam original ↗

17 Jan 2026

2025

Entropic Instruction Following: Does Semantic Coherence Help LLMs Follow Instructions?original ↗

2 Dec 2025

Elements Of Mechanistic Interpretability: From Observation to Causation original ↗

26 Oct 2025