Why winning training recipes at proxy scale can lose at target scale
gerardsans · x · 2026-09-10
A reply to BlackHC's educational thread on Bayesian model selection and scaling laws: three criteria for "best", when they disagree under scaling, and what that means for evals. The replier argues current recipe-scaling has blindspots — it ignores token isomorphism structure, treats curation and loss as generic ML problems, and that transformers learn co-occurrence geometry under repetition. Without distribution support in the frozen pretraining landscape, scaffolding and prettier losses can't fill interpolation voids; loss only scores next-token fit on the training measure.
More from Research
- Open-source decision models sold their own base in Red Alert 2 — then learned until no human could beat them — heathercmiller · 2026-10-27
- DeepMind paper: LLMs can't sample randomly on their own — pass random numbers from code instead — rohanpaul_ai · 2026-10-27
- World Model Assigns Weight and Friction to Objects, Then Simulates with a Physics Engine — jonstephens85 · 2026-10-27
- Token Collapse: Study Finds Graph Soft Prompting Tokens in LLMs Fail to Encode Structure — Sapienza · 2026-10-27
- Mathathon 2.0 launches with Terence Tao's SAIR, $1.5M in compute for teams — _sathvikr · 2026-10-27
- The 'sweet lesson' is real: two kinds of ML problems need opposite strategies — mike64_t · 2026-10-27