18 lines
1.1 KiB
Markdown
18 lines
1.1 KiB
Markdown
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.
|
|
---
|
|
license: cc-by-nc-4.0
|
|
---
|
|
# [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_camel-ai__CAMEL-13B-Role-Playing-Data)
|
|
|
|
| Metric | Value |
|
|
|-----------------------|---------------------------|
|
|
| Avg. | 45.03 |
|
|
| ARC (25-shot) | 54.95 |
|
|
| HellaSwag (10-shot) | 79.25 |
|
|
| MMLU (5-shot) | 46.61 |
|
|
| TruthfulQA (0-shot) | 46.35 |
|
|
| Winogrande (5-shot) | 74.03 |
|
|
| GSM8K (5-shot) | 7.35 |
|
|
| DROP (3-shot) | 6.66 |
|