This is a merge of pre-trained language models created using
mergekit.
We adapted the code from
FastChat to benchmark our model with GPT-4 as a judge. Here is the result
12|--------|-----------------------------------------|------|----------|
3| First | tlphams/Wizard-Zephyr-Orpo-8x22B | 1 | 9.1625 |
4| | mistralai/Mixtral-8x22B-Instruct-v0.1 | 1 | 9.1500 |
5| Second | tlphams/Wizard-Zephyr-Orpo-8x22B | 2 | 8.873418 |
6| | mistralai/Mixtral-8x22B-Instruct-v0.1 | 2 | 8.250000 |
7| Average| tlphams/Wizard-Zephyr-Orpo-8x22B | | 9.018868 |
8| | mistralai/Mixtral-8x22B-Instruct-v0.1 | | 8.700000 |
The score is slightly lower than
alpindale/WizardLM-2-8x22B, but still higher than GPT-4-0314. Then the research and experimental work still need to continue ^^
