Update README.md
Browse files
README.md
CHANGED
@@ -1,4 +1,21 @@
|
|
1 |
---
|
2 |
license: llama2
|
3 |
---
|
4 |
-
llama-2-70b-instruct-v1.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: llama2
|
3 |
---
|
4 |
+
llama-2-70b-instruct-v1.1
|
5 |
+
|
6 |
+
* Model license: Llama 2 Community License Agreement
|
7 |
+
* Basic usage: [notebook](assets/basic_inference_llama_2_dolphin.ipynb)
|
8 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
9 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_dfurman__llama-2-70b-dolphin-peft)
|
10 |
+
|
11 |
+
| Metric | Value |
|
12 |
+
|-----------------------|---------------------------|
|
13 |
+
| Avg. | 65.72 |
|
14 |
+
| ARC (AI2) (25-shot) | 69.62 |
|
15 |
+
| HellaSwag (10-shot) | 86.82 |
|
16 |
+
| MMLU (5-shot) | 69.18 |
|
17 |
+
| TruthfulQA (0-shot) | 57.43 |
|
18 |
+
| Winogrande (5-shot) | 83.9 |
|
19 |
+
| GSM8K (5-shot) | 27.37
|
20 |
+
|
21 |
+
- PEFT 0.5.0.dev0
|