Adding Evaluation Results
#4
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -80,4 +80,17 @@ All airoboros models and datasets are intended and licensed for research use onl
|
|
| 80 |
- the base model is LLaMa, which has it's own special research license
|
| 81 |
- 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
|
| 82 |
|
| 83 |
-
So, to reiterate: this model (and datasets) cannot be used commercially.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
- the base model is LLaMa, which has it's own special research license
|
| 81 |
- 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
|
| 82 |
|
| 83 |
+
So, to reiterate: this model (and datasets) cannot be used commercially.
|
| 84 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 85 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_jondurbin__airoboros-7b-gpt4-1.2)
|
| 86 |
+
|
| 87 |
+
| Metric | Value |
|
| 88 |
+
|-----------------------|---------------------------|
|
| 89 |
+
| Avg. | 43.75 |
|
| 90 |
+
| ARC (25-shot) | 52.13 |
|
| 91 |
+
| HellaSwag (10-shot) | 78.14 |
|
| 92 |
+
| MMLU (5-shot) | 38.64 |
|
| 93 |
+
| TruthfulQA (0-shot) | 41.79 |
|
| 94 |
+
| Winogrande (5-shot) | 71.67 |
|
| 95 |
+
| GSM8K (5-shot) | 2.12 |
|
| 96 |
+
| DROP (3-shot) | 21.76 |
|