mixtral-ties / README.md
monology's picture
Update README.md
921017c verified
---
base_model:
- monology/mixtral-slerp0
- monology/mixtral-slerp5
- monology/mixtral-slerp6
- monology/mixtral-slerp2
- mistralai/Mistral-7B-v0.1
- monology/mixtral-slerp3
- monology/mixtral-slerp7
- monology/mixtral-slerp4
- monology/mixtral-slerp1
library_name: transformers
tags:
- mergekit
- merge
license: apache-2.0
---
# mixtral-ties
For experimental purposes. Probably not very good.
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) as a base.
### Models Merged
The following models were included in the merge:
* [monology/mixtral-slerp0](https://huggingface.co/monology/mixtral-slerp0)
* [monology/mixtral-slerp5](https://huggingface.co/monology/mixtral-slerp5)
* [monology/mixtral-slerp6](https://huggingface.co/monology/mixtral-slerp6)
* [monology/mixtral-slerp2](https://huggingface.co/monology/mixtral-slerp2)
* [monology/mixtral-slerp3](https://huggingface.co/monology/mixtral-slerp3)
* [monology/mixtral-slerp7](https://huggingface.co/monology/mixtral-slerp7)
* [monology/mixtral-slerp4](https://huggingface.co/monology/mixtral-slerp4)
* [monology/mixtral-slerp1](https://huggingface.co/monology/mixtral-slerp1)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: mistralai/Mistral-7B-v0.1
# no parameters necessary for base model
- model: monology/mixtral-slerp0
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp1
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp2
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp3
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp4
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp5
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp6
parameters:
density: 0.5
weight: 0.1
- model: monology/mixtral-slerp7
parameters:
density: 0.5
weight: 0.1
merge_method: ties
base_model: mistralai/Mistral-7B-v0.1
parameters:
normalize: true
dtype: float16
```