Adding Evaluation Results
#1
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -31,4 +31,17 @@ Baize (白泽) is a mythical creature in Chinese folklore, who speaks human lang
|
|
31 |
- [Quora Dialogs](https://github.com/project-baize/baize) (54,456):
|
32 |
- [Medical Dialogs](https://github.com/project-baize/baize) (46,867)
|
33 |
|
34 |
-
More details can be found in the Baize [GitHub]((https://github.com/project-baize/baize))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
- [Quora Dialogs](https://github.com/project-baize/baize) (54,456):
|
32 |
- [Medical Dialogs](https://github.com/project-baize/baize) (46,867)
|
33 |
|
34 |
+
More details can be found in the Baize [GitHub]((https://github.com/project-baize/baize))
|
35 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
36 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_project-baize__baize-healthcare-lora-7B)
|
37 |
+
|
38 |
+
| Metric | Value |
|
39 |
+
|-----------------------|---------------------------|
|
40 |
+
| Avg. | 41.66 |
|
41 |
+
| ARC (25-shot) | 54.1 |
|
42 |
+
| HellaSwag (10-shot) | 77.32 |
|
43 |
+
| MMLU (5-shot) | 37.09 |
|
44 |
+
| TruthfulQA (0-shot) | 39.96 |
|
45 |
+
| Winogrande (5-shot) | 72.85 |
|
46 |
+
| GSM8K (5-shot) | 4.4 |
|
47 |
+
| DROP (3-shot) | 5.93 |
|