finding
active
finding:olmo-models-trained-with-larger-injection-gaps-n-1-10-20-50-100-reach-higher-task-loss-despite-equal-global-frequency

OLMo models trained with larger injection gaps N∈{1,10,20,50,100} reach higher task loss despite equal global frequency

Fig. 6b.

Source paper

extracted_from
Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task Retention
(2026) · Jing Huang · Daniel Wurgaft · Rachit Bansal · Laura Ruis +5

Neighborhood — ranked by edge-count