42 lines
1.0 KiB
Markdown
42 lines
1.0 KiB
Markdown
|
---
|
||
|
|
base_model:
|
||
|
|
- Epiculous/Fett-uccine-7B
|
||
|
|
library_name: transformers
|
||
|
|
tags:
|
||
|
|
- mergekit
|
||
|
|
- merge
|
||
|
|
license: agpl-3.0
|
||
|
|
---
|
||
|
|
# merge
|
||
|
|
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
||
|
|
|
||
|
|
## Merge Details
|
||
|
|
**Scores lower than the original model! Not recommended unless you're experimenting.**
|
||
|
|
|
||
|
|
|
||
|
|
This model is a experiment of using passthrough on 7b models to further merge them with 10.7b/11b models for fun. i doubt there will be any benfits of this model over the orignal.
|
||
|
|
### Merge Method
|
||
|
|
|
||
|
|
This model was merged using the passthrough merge method.
|
||
|
|
|
||
|
|
### Models Merged
|
||
|
|
|
||
|
|
The following models were included in the merge:
|
||
|
|
* [Epiculous/Fett-uccine-7B](https://huggingface.co/Epiculous/Fett-uccine-7B)
|
||
|
|
|
||
|
|
### Configuration
|
||
|
|
|
||
|
|
The following YAML configuration was used to produce this model:
|
||
|
|
|
||
|
|
```yaml
|
||
|
|
slices:
|
||
|
|
- sources:
|
||
|
|
- model: Epiculous/Fett-uccine-7B
|
||
|
|
layer_range: [0, 24]
|
||
|
|
- sources:
|
||
|
|
- model: Epiculous/Fett-uccine-7B
|
||
|
|
layer_range: [8, 32]
|
||
|
|
merge_method: passthrough
|
||
|
|
dtype: float16
|
||
|
|
|
||
|
|
```
|