51 lines
1.3 KiB
Markdown
51 lines
1.3 KiB
Markdown
|
---
|
||
|
|
base_model: []
|
||
|
|
library_name: transformers
|
||
|
|
tags:
|
||
|
|
- mergekit
|
||
|
|
- merge
|
||
|
|
|
||
|
|
---
|
||
|
|
# Qwen3-8B-PDAPT-SLERP
|
||
|
|
|
||
|
|
## PARTAGES
|
||
|
|
|
||
|
|
The PARTAGES project is a French national initiative aiming to develop specialized language models for use in the automation of document-processing tasks in the French healthcare system.
|
||
|
|
All foundational PARTAGES data and models will be released as open-source tools.
|
||
|
|
|
||
|
|
**Note:** While these models have been developed in the context of a healthcare-focused project, they are intended for research use only and should not under any circumstances be used as-is in production settings or as a replacement for medical advice of any kind.
|
||
|
|
|
||
|
|
|
||
|
|
## Merge Details
|
||
|
|
|
||
|
|
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
||
|
|
|
||
|
|
### Merge Method
|
||
|
|
|
||
|
|
This model was merged using the [SLERP](https://en.wikipedia.org/wiki/Slerp) merge method.
|
||
|
|
|
||
|
|
### Models Merged
|
||
|
|
|
||
|
|
The following models were included in the merge:
|
||
|
|
* PARTAGES/Qwen3-8B-Base-PARTAGES-v2-1440
|
||
|
|
* Qwen/Qwen3-8B-Base
|
||
|
|
|
||
|
|
### Configuration
|
||
|
|
|
||
|
|
The following YAML configuration was used to produce this model:
|
||
|
|
|
||
|
|
```yaml
|
||
|
|
slices:
|
||
|
|
- sources:
|
||
|
|
- model: Qwen/Qwen3-8B-Base
|
||
|
|
layer_range: [0, 36]
|
||
|
|
- model: [ANON]
|
||
|
|
layer_range: [0, 36]
|
||
|
|
merge_method: slerp
|
||
|
|
base_model: Qwen/Qwen3-8B-Base
|
||
|
|
parameters:
|
||
|
|
t:
|
||
|
|
- value: 0.5
|
||
|
|
dtype: bfloat16
|
||
|
|
```
|