flâneur — a map of the web's best reading

LoRA Without Regret - Thinking Machines Lab

thinkingmachines.ai · 5,825 words · saved by 32 readers

How LoRA matches full training performance more broadly than expected.

Today's leading language models contain upwards of a trillion parameters, pretrained on tens of trillions of tokens. Base model performance keeps improving with scale, as these trillions are necessary for learning and representing all the patterns in written-down human knowledge. In contrast, post-training involves smaller datasets and generally focuses on narrower domains of knowledge and ranges of behavior. It seems wasteful to use a terabit of weights to represent updates from a gigabit or megabit of training data. This intuition has motivated parameter efficient fine-tuning (PEFT), which a

Explore this link on the map →

saved by

related reading