The code is most likely released after the ICLR25 submission deadline. Sorry for the inconvenience.
If you are interested in the finetuning tasks, we basically use the same finetuning code at https://github.com/BaohaoLiao/ApiQ.
[NeurIPS 2024] 3-in-1: 2D Rotary Adaptation for Efficient Finetuning, Efficient Batching and Composability
The code is most likely released after the ICLR25 submission deadline. Sorry for the inconvenience.
If you are interested in the finetuning tasks, we basically use the same finetuning code at https://github.com/BaohaoLiao/ApiQ.
[NeurIPS 2024] 3-in-1: 2D Rotary Adaptation for Efficient Finetuning, Efficient Batching and Composability
https://arxiv.org/abs/2409.00119
MIT License