Adding Evaluation Results
#1
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -94,3 +94,17 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 94 |
|
| 95 |
|
| 96 |
- PEFT 0.5.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
|
| 95 |
|
| 96 |
- PEFT 0.5.0
|
| 97 |
+
|
| 98 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 99 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_uukuguy__Mistral-7B-OpenOrca-lora)
|
| 100 |
+
|
| 101 |
+
| Metric | Value |
|
| 102 |
+
|-----------------------|---------------------------|
|
| 103 |
+
| Avg. | 50.72 |
|
| 104 |
+
| ARC (25-shot) | 61.95 |
|
| 105 |
+
| HellaSwag (10-shot) | 83.62 |
|
| 106 |
+
| MMLU (5-shot) | 64.16 |
|
| 107 |
+
| TruthfulQA (0-shot) | 42.74 |
|
| 108 |
+
| Winogrande (5-shot) | 79.08 |
|
| 109 |
+
| GSM8K (5-shot) | 17.29 |
|
| 110 |
+
| DROP (3-shot) | 6.19 |
|