Adding Evaluation Results
#6
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -44,4 +44,17 @@ A preliminary evaluation of the model quality is conducted by creating a set of
|
|
44 |
|
45 |
## Major updates of weights v1.1
|
46 |
- Refactor the tokenization and separator. In Vicuna v1.1, the separator has been changed from `"###"` to the EOS token `"</s>"`. This change makes it easier to determine the generation stop criteria and enables better compatibility with other libraries.
|
47 |
-
- Fix the supervised fine-tuning loss computation for better model quality.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
## Major updates of weights v1.1
|
46 |
- Refactor the tokenization and separator. In Vicuna v1.1, the separator has been changed from `"###"` to the EOS token `"</s>"`. This change makes it easier to determine the generation stop criteria and enables better compatibility with other libraries.
|
47 |
+
- Fix the supervised fine-tuning loss computation for better model quality.
|
48 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
49 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_eachadea__vicuna-7b-1.1)
|
50 |
+
|
51 |
+
| Metric | Value |
|
52 |
+
|-----------------------|---------------------------|
|
53 |
+
| Avg. | 45.6 |
|
54 |
+
| ARC (25-shot) | 53.67 |
|
55 |
+
| HellaSwag (10-shot) | 77.46 |
|
56 |
+
| MMLU (5-shot) | 45.63 |
|
57 |
+
| TruthfulQA (0-shot) | 48.94 |
|
58 |
+
| Winogrande (5-shot) | 70.96 |
|
59 |
+
| GSM8K (5-shot) | 5.53 |
|
60 |
+
| DROP (3-shot) | 16.98 |
|