Adding Evaluation Results
#3
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -111,4 +111,17 @@ Llama 2 is licensed under the LLAMA 2 Community License, Copyright (c) Meta Plat
|
|
111 |
archivePrefix={arXiv},
|
112 |
primaryClass={cs.CL}
|
113 |
}
|
114 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
111 |
archivePrefix={arXiv},
|
112 |
primaryClass={cs.CL}
|
113 |
}
|
114 |
+
```
|
115 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
116 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_elyza__ELYZA-japanese-Llama-2-7b-fast-instruct)
|
117 |
+
|
118 |
+
| Metric | Value |
|
119 |
+
|-----------------------|---------------------------|
|
120 |
+
| Avg. | 42.96 |
|
121 |
+
| ARC (25-shot) | 53.75 |
|
122 |
+
| HellaSwag (10-shot) | 77.55 |
|
123 |
+
| MMLU (5-shot) | 46.85 |
|
124 |
+
| TruthfulQA (0-shot) | 38.84 |
|
125 |
+
| Winogrande (5-shot) | 71.59 |
|
126 |
+
| GSM8K (5-shot) | 6.29 |
|
127 |
+
| DROP (3-shot) | 5.84 |
|