license: apache-2.0 | |
tags: | |
- merge | |
- mergekit | |
- lazymergekit | |
- microsoft/Phi-3.5-mini-instruct | |
- microsoft/Phi-3-mini-4k-instruct | |
- microsoft/Phi-3-mini-128k-instruct | |
# phi-merge-arithmetic | |
phi-merge-arithmetic is a merge of the following models using [mergekit](https://github.com/cg123/mergekit): | |
* [microsoft/Phi-3.5-mini-instruct](https://huggingface.co/microsoft/Phi-3.5-mini-instruct) | |
* [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) | |
* [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) | |
## 🧩 Configuration | |
```yaml | |
models: | |
- model: microsoft/Phi-3.5-mini-instruct | |
parameters: | |
weight: 1.0 | |
- model: microsoft/Phi-3-mini-4k-instruct | |
parameters: | |
weight: 0.6 | |
- model: microsoft/Phi-3-mini-128k-instruct | |
parameters: | |
weight: 0.2 | |
base_model: microsoft/Phi-3.5-mini-instruct | |
merge_method: task_arithmetic | |
dtype: float16 | |
``` |