megasalad-2 / README.md
appvoid's picture
Upload folder using huggingface_hub
db90696 verified
---
base_model:
- h2oai/h2o-danube3-500m-base
- appvoid/text-arco
- appvoid/arco-2
- Fizzarolli/clite-500m
- appvoid/massive
library_name: transformers
tags:
- mergekit
- merge
---
# merge
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 [h2oai/h2o-danube3-500m-base](https://huggingface.co/h2oai/h2o-danube3-500m-base) as a base.
### Models Merged
The following models were included in the merge:
* [appvoid/text-arco](https://huggingface.co/appvoid/text-arco)
* [appvoid/arco-2](https://huggingface.co/appvoid/arco-2)
* [Fizzarolli/clite-500m](https://huggingface.co/Fizzarolli/clite-500m)
* [appvoid/massive](https://huggingface.co/appvoid/massive)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: h2oai/h2o-danube3-500m-base
#no parameters necessary for base model
- model: appvoid/text-arco
parameters:
density: 0.5
weight: 0.25
- model: appvoid/arco-2
parameters:
density: 0.5
weight: 0.25
- model: appvoid/massive
parameters:
density: 0.5
weight: 0.25
- model: Fizzarolli/clite-500m
parameters:
density: 0.5
weight: 0.25
merge_method: ties
base_model: h2oai/h2o-danube3-500m-base
parameters:
normalize: false
int8_mask: true
dtype: float16
```