Pith. sign in

Nelder-Mead

In thepostnorm experiment, they were as follows: for 4M effective parameters, 3% at checkpoint 1, 4% at 2, 5% at 3, 6% at 4, 7% at 5–8, for 34M, 134M, 7% at 1–2, 8% at 3, 10% at 4–8 · 2023

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
browse 1 citing papers

fields

cs.LG 1

years

2026 1

verdicts

CONDITIONAL 1

representative citing papers

Small-Scale Experiments: Are We There Yet?

cs.LG · 2026-08-12 · conditional · novelty 7.0

With about 256 hyperparameter configurations per scale, scaling laws emerge at 4M parameters, and the apparent small-scale unreliability is largely a hyperparameter-tuning artifact.

citing papers explorer

Showing 1 of 1 citing paper.

  • Small-Scale Experiments: Are We There Yet? cs.LG · 2026-08-12 · conditional · none · ref 8

    With about 256 hyperparameter configurations per scale, scaling laws emerge at 4M parameters, and the apparent small-scale unreliability is largely a hyperparameter-tuning artifact.