Knowledge Distillation of Large Language Models
Paper • 2306.08543 • Published • 23
teacher-Llama-13B is a Llama-13B model supervised fine-tuned on databricks-dolly-15k.
It is used as the teacher model for MiniLLM series.
@inproceedings{minillm,
title={MiniLLM: Knowledge Distillation of Large Language Models},
author={Gu, Yuxian and Dong, Li and Wei, Furu and Huang, Minlie},
booktitle={Proceedings of ICLR},
year={2024}
}