该模型使用bloom-7b1，使用百万中英文指令数据，进行指令微调。

更多详情见[Firefly项目](https://github.com/yangjianxin1/Firefly)
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_YeungNLP__firefly-bloom-7b1)

| Metric                | Value                     |
|-----------------------|---------------------------|
| Avg.                  | 34.99   |
| ARC (25-shot)         | 40.44          |
| HellaSwag (10-shot)   | 61.2    |
| MMLU (5-shot)         | 26.83         |
| TruthfulQA (0-shot)   | 40.83   |
| Winogrande (5-shot)   | 64.56   |
| GSM8K (5-shot)        | 0.68        |
| DROP (3-shot)         | 10.37         |