Adding Evaluation Results
#1
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -106,4 +106,17 @@ All airoboros models and datasets are intended and licensed for research use onl
|
|
106 |
- the base model is LLaMa, which has it's own special research license
|
107 |
- the dataset(s) were generated with OpenAI (gpt-4 and/or gpt-3.5-turbo), which has a clausing saying the data can't be used to create models to compete with openai
|
108 |
|
109 |
-
So, to reiterate: this model (and datasets) cannot be used commercially.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
106 |
- the base model is LLaMa, which has it's own special research license
|
107 |
- the dataset(s) were generated with OpenAI (gpt-4 and/or gpt-3.5-turbo), which has a clausing saying the data can't be used to create models to compete with openai
|
108 |
|
109 |
+
So, to reiterate: this model (and datasets) cannot be used commercially.
|
110 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
111 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_jondurbin__airoboros-65b-gpt4-1.3)
|
112 |
+
|
113 |
+
| Metric | Value |
|
114 |
+
|-----------------------|---------------------------|
|
115 |
+
| Avg. | 58.57 |
|
116 |
+
| ARC (25-shot) | 66.13 |
|
117 |
+
| HellaSwag (10-shot) | 85.99 |
|
118 |
+
| MMLU (5-shot) | 63.89 |
|
119 |
+
| TruthfulQA (0-shot) | 51.32 |
|
120 |
+
| Winogrande (5-shot) | 79.95 |
|
121 |
+
| GSM8K (5-shot) | 13.65 |
|
122 |
+
| DROP (3-shot) | 49.07 |
|