Concept
Chinchilla scaling laws
Chinchilla scaling laws describe the compute-optimal relationship between model size and training data, published by DeepMind in 2022 (the "Chinchilla" paper). They correct an earlier assumption from the 2020 Kaplan scaling laws, which had steered practitioners toward making…
The rest of “Chinchilla scaling laws” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→