Adding Evaluation Results
#5
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -141,4 +141,17 @@ Beware of hallucinations: Outputs are often factually wrong or misleading.
|
|
| 141 |
Replies might look convincing (at first glance) while containing completely
|
| 142 |
made up false statements.
|
| 143 |
|
| 144 |
-
This model is usable only for English conversations.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 141 |
Replies might look convincing (at first glance) while containing completely
|
| 142 |
made up false statements.
|
| 143 |
|
| 144 |
+
This model is usable only for English conversations.
|
| 145 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 146 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_amazon__LightGPT)
|
| 147 |
+
|
| 148 |
+
| Metric | Value |
|
| 149 |
+
|-----------------------|---------------------------|
|
| 150 |
+
| Avg. | 35.49 |
|
| 151 |
+
| ARC (25-shot) | 39.93 |
|
| 152 |
+
| HellaSwag (10-shot) | 63.82 |
|
| 153 |
+
| MMLU (5-shot) | 28.45 |
|
| 154 |
+
| TruthfulQA (0-shot) | 36.69 |
|
| 155 |
+
| Winogrande (5-shot) | 64.48 |
|
| 156 |
+
| GSM8K (5-shot) | 3.87 |
|
| 157 |
+
| DROP (3-shot) | 11.19 |
|