Is More Data Worth the Cost? Dataset Scaling Laws in a Tiny Attention-Only Decoder
Type
conference poster
Date Issued
2026-04-10
Author(s)
;
;
Rico Stödeli
;
;
;
Abstract
Training Transformer language models is expensive, as performance typically improves with increasing dataset size and computational budget. Although scaling laws describe this trend at large scale, their implications in controlled, smaller-scale settings remain less explored. In this work, we isolate dataset-size effects using a strongly reduced attention-only decoder architecture. By training on progressively larger power-of-two subsets, we observe smooth performance improvements accompanied by clear diminishing returns, consistent with scaling-law behavior. Using only about 30% of the training data is sufficient to reach approximately 90% of the full-data validation token-level accuracy. These results provide actionable insights into dataset scaling in a controlled, component-isolated setting and offer practical guidance for balancing dataset size and computational cost in compute- and data-restricted environments, such as small research labs and exploratory model development.
Language
English
HSG Classification
contribution to scientific community
Refereed
Yes
Event Title
International Conference on Learning Representations (ICLR 2026) - DATA-FM Workshop
Event Location
Rio de Janeiro
Official URL
Subject(s)
File(s)![Thumbnail Image]()
Name
2604.09389v1.pdf
Size
2.54 MB
Format
Adobe PDF
Checksum (MD5)
b72df9930ef6429f5f5b849314812022