Papers › Problems with Chinchilla Approach 2: Systematic Biases in IsoFLOP Parabola Fits

Problems with Chinchilla Approach 2: Systematic Biases in IsoFLOP Parabola Fits

21 Mar 2026arXiv:2603.22339added by Syntology

Eric Czech, Zhiwei Xu, Yael Elmatad, Yixin Wang, William Held

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

Chinchilla Approach 2 is among the most widely used methods for fitting neural scaling laws. Its parabolic approximation introduces systematic biases in compute-optimal allocation estimates, even on noise-free synthetic data. Applied to published Llama 3 IsoFLOP data at open frontier compute scales, these biases imply a parameter underallocation corresponding to 6.5% of the 3.8×10²⁵ FLOP training budget and $1.4M (90% CI: $412K-$2.9M) in unnecessary compute at 50% H100 MFU. Simulated multimodal model misallocations show even greater opportunity costs due to higher loss surface asymmetry. Three sources of this error are examined: IsoFLOP sampling grid width (Taylor approximation accuracy), uncentered IsoFLOP sampling, and loss surface asymmetry (α≠β). Chinchilla Approach 3 largely eliminates these biases but is often regarded as less data-efficient, numerically unstable, prone to local minima, and harder to implement. Each concern is shown to be unfounded or addressable, especially when the partially linear structure of the objective is exploited via Variable Projection, enabling unbiased inference on all five loss surface parameters through a two-dimensional optimization that is well-conditioned, analytically differentiable, and amenable to dense, or even exhaustive, grid search. It may serve as a more convenient replacement for Approach 2 or a more scalable alternative for adaptations of Approach 3 to richer scaling law formulations. See https://github.com/Open-Athena/vpnls for details and https://openathena.ai/scaling-law-analysis for other results from this study.

Code

Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Syntology holds the repository link but has not harvested or run code from it.

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2603.22339, was issued in March 2026, after that date, so the archive has no leaderboard rows for it.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections