Adding Evaluation Results
#4
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -587,3 +587,17 @@ I am purposingly leaving this license ambiguous (other than the fact you must co
|
|
| 587 |
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
|
| 588 |
|
| 589 |
Either way, by using this model, you agree to completely indemnify me.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 587 |
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
|
| 588 |
|
| 589 |
Either way, by using this model, you agree to completely indemnify me.
|
| 590 |
+
|
| 591 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 592 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_TheBloke__Airoboros-L2-70B-2.1-GPTQ)
|
| 593 |
+
|
| 594 |
+
| Metric | Value |
|
| 595 |
+
|-----------------------|---------------------------|
|
| 596 |
+
| Avg. | 61.76 |
|
| 597 |
+
| ARC (25-shot) | 70.39 |
|
| 598 |
+
| HellaSwag (10-shot) | 86.54 |
|
| 599 |
+
| MMLU (5-shot) | 68.89 |
|
| 600 |
+
| TruthfulQA (0-shot) | 55.55 |
|
| 601 |
+
| Winogrande (5-shot) | 81.61 |
|
| 602 |
+
| GSM8K (5-shot) | 15.24 |
|
| 603 |
+
| DROP (3-shot) | 54.1 |
|