Edit model card

We are glad to introduce the PandaLM version of Alpaca. To highlight the effectiveness of using PandaLM-7B for instruction tuning LLMs, we check the performance of models tuned with PandaLM’s selected optimal hyperparameters. This version and original alpaca version have been submitted to hugging face Open LLM Leaderboard.

the full checkpoint has been uploaded to the Hugging face, so you can just load the model and tokenizer for downstream tasks.

from transformers import AutoTokenizer,AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("WeOpenML/PandaLM-Alpaca-7B-v1",use_fast=False)
model = AutoModelForCausalLM.from_pretrained("WeOpenML/PandaLM-Alpaca-7B-v1")

For more information about PandaLM, pls check out our github, our paper and PandaLM model. The repo is under Apache License 2.0.

Updates


  • 2023.7.21: We updated the model card and basic info.
  • 2023.7.18: We released the checkpoint on the Hugging face.

Citation

@misc{pandalm2023,
      title={PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization}, 
      author={Wang, Yidong and Yu, Zhuohao and Zeng, Zhengran and Yang, Linyi and Wang, Cunxiang and Chen, Hao and Jiang, Chaoya and Xie, Rui and Wang, Jindong and Xie, Xing and Ye, Wei and Zhang, Shikun and Zhang, Yue},
      year={2023},
      journal={arXiv preprint arXiv:2306.05087}
}

@misc{PandaLM,
  author = {Wang, Yidong and Yu, Zhuohao and Zeng, Zhengran and Yang, Linyi and Heng, Qiang and Wang, Cunxiang and Chen, Hao and Jiang, Chaoya and Xie, Rui and Wang, Jindong and Xie, Xing and Ye, Wei and Zhang, Shikun and Zhang, Yue},
  title = {PandaLM: Reproducible and Automated Language Model Assessment},
  year = {2023},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/WeOpenML/PandaLM}},
}
Downloads last month
1,442
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Spaces using WeOpenML/PandaLM-Alpaca-7B-v1 24