conftrace_

loss trajectory

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

gradient error (1) gradient clustering (1) large language model (1) training trajectory (1) training trajectories (1) mathematical problem solving (1) data efficiency (1) supervised fine-tuning (1) data selection (1)

Papers

SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models NeurIPS 2024