LISA: Layerwise Importance Sampling for Memory-Efficient LLM Fine-Tuning Posted by convexstrictly on Wednesday March 27, 2024 @01:17PM from the arxiv.org dept. https://arxiv.org/abs/2403.17919 ( 1 comment | 2 points | on HN )
https://twitter.com/Rui45898440/status/1772996453557997924