Adding Evaluation Results
#2
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -20,3 +20,17 @@ The base OPT-6.7B model is licensed under the OPT-175B license, Copyright (c) Me
|
|
20 |
No formal evaluation is available for this model at this time. This blend was created in FP16, due to available memory constraints.
|
21 |
|
22 |
It is recommend to use this model with the KoboldAI software. All feedback and comments can be directed to Concedo on the KoboldAI discord.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
No formal evaluation is available for this model at this time. This blend was created in FP16, due to available memory constraints.
|
21 |
|
22 |
It is recommend to use this model with the KoboldAI software. All feedback and comments can be directed to Concedo on the KoboldAI discord.
|
23 |
+
|
24 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
25 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_KoboldAI__OPT-6.7B-Nerybus-Mix)
|
26 |
+
|
27 |
+
| Metric | Value |
|
28 |
+
|-----------------------|---------------------------|
|
29 |
+
| Avg. | 33.98 |
|
30 |
+
| ARC (25-shot) | 39.16 |
|
31 |
+
| HellaSwag (10-shot) | 68.63 |
|
32 |
+
| MMLU (5-shot) | 24.47 |
|
33 |
+
| TruthfulQA (0-shot) | 34.84 |
|
34 |
+
| Winogrande (5-shot) | 65.11 |
|
35 |
+
| GSM8K (5-shot) | 0.76 |
|
36 |
+
| DROP (3-shot) | 4.85 |
|