metadata
license: apache-2.0
datasets:
- databricks/databricks-dolly-15k
language:
- en
metrics:
- rouge
base_model:
- facebook/opt-2.7B
pipeline_tag: text-generation
init-OPT-2.7B
init-OPT-2.7B is an OPT-2.7B model supervised fine-tuned on databricks-dolly-15k.
It is used as the initialization for training MiniLLM.
Citation
@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}
}