Adding Evaluation Results
#2
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -29,4 +29,17 @@ I would like to thank AutoMeta for providing me with the computing power necessa
|
|
| 29 |
|
| 30 |
### Changes
|
| 31 |
This is the second version of the 3B parameter Guanaco uncensored model.
|
| 32 |
-
The model has been fine-tuned on the V2 of the Guanaco unfiltered dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
### Changes
|
| 31 |
This is the second version of the 3B parameter Guanaco uncensored model.
|
| 32 |
+
The model has been fine-tuned on the V2 of the Guanaco unfiltered dataset.
|
| 33 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 34 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Fredithefish__Guanaco-3B-Uncensored-v2)
|
| 35 |
+
|
| 36 |
+
| Metric | Value |
|
| 37 |
+
|-----------------------|---------------------------|
|
| 38 |
+
| Avg. | 34.35 |
|
| 39 |
+
| ARC (25-shot) | 42.15 |
|
| 40 |
+
| HellaSwag (10-shot) | 66.72 |
|
| 41 |
+
| MMLU (5-shot) | 26.18 |
|
| 42 |
+
| TruthfulQA (0-shot) | 35.21 |
|
| 43 |
+
| Winogrande (5-shot) | 63.3 |
|
| 44 |
+
| GSM8K (5-shot) | 0.3 |
|
| 45 |
+
| DROP (3-shot) | 6.6 |
|