CAMEL-13B-Role-Playing-Data is a chat large language model obtained by finetuning LLaMA-13B model on a total of 229K conversations created through our role-playing framework proposed in [CAMEL](https://arxiv.org/abs/2303.17760). We evaluate our model offline using EleutherAI's language model evaluation harness used by Huggingface's Open LLM Benchmark. CAMEL-13B scores an average of **57.2**, outperfroming LLaMA-30B (58.3)!

| Model       | size | ARC-C  (25 shots, acc_norm) | HellaSwag  (10 shots, acc_norm) | MMLU  (5 shots, acc_norm) | TruthfulQA  (0 shot, mc2) | Average | Delta |
|-------------|:----:|:---------------------------:|:-------------------------------:|:-------------------------:|:-------------------------:|:-------:|-------|
| LLaMA       |  13B |             50.8            |               78.9              |            37.7           |            39.9           |   51.8  |   -   |
| Vicuna      |  13B |             47.4            |               75.2              |            39.6           |            49.8           |   53.7  |  1.9  |
| CAMEL       |  13B |             54.9            |               79.3              |            48.5           |            46.2           |   57.2  |  5.4  |
| LLaMA       |  30B |             57.1            |               82.6              |            45.7           |            42.3           |   56.9  |  5.1  |

---
license: cc-by-nc-4.0
---