Update README.md
Browse files
README.md
CHANGED
@@ -24,16 +24,3 @@ GGML quants available [here](https://huggingface.co/TheBloke/Puma-3b-GGML).</br>
|
|
24 |
GPTQ quants available [here](https://huggingface.co/TheBloke/Puma-3b-GPTQ).
|
25 |
|
26 |
Note: Don't expect this model to be good, I was just starting out to finetune. So don't roast me please!
|
27 |
-
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
28 |
-
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_acrastt__Puma-3B)
|
29 |
-
|
30 |
-
| Metric | Value |
|
31 |
-
|-----------------------|---------------------------|
|
32 |
-
| Avg. | 35.93 |
|
33 |
-
| ARC (25-shot) | 41.3 |
|
34 |
-
| HellaSwag (10-shot) | 71.85 |
|
35 |
-
| MMLU (5-shot) | 27.51 |
|
36 |
-
| TruthfulQA (0-shot) | 38.34 |
|
37 |
-
| Winogrande (5-shot) | 66.38 |
|
38 |
-
| GSM8K (5-shot) | 0.76 |
|
39 |
-
| DROP (3-shot) | 5.38 |
|
|
|
24 |
GPTQ quants available [here](https://huggingface.co/TheBloke/Puma-3b-GPTQ).
|
25 |
|
26 |
Note: Don't expect this model to be good, I was just starting out to finetune. So don't roast me please!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|