|
|
--- |
|
|
license: cc-by-nc-4.0 |
|
|
base_model: |
|
|
- knifeayumu/Cydonia-v1.2-Magnum-v4-22B |
|
|
- Steelskull/MSM-MS-Cydrion-22B |
|
|
library_name: transformers |
|
|
tags: |
|
|
- mergekit |
|
|
- merge |
|
|
|
|
|
--- |
|
|
# What is this? |
|
|
|
|
|
Another 22B model. Decent, I think? I tested with Q4_K_S, so can't tell the real performance of it. The choice is your! Enjoy! |
|
|
|
|
|
Template: Mistral, specific is Mistral V3, don't use V3-Tekken. If you occurs of model start talking for you, you should use ChatML. |
|
|
|
|
|
|
|
|
<details> |
|
|
<summary>Merge Detail</summary> |
|
|
<p> |
|
|
### Models Merged |
|
|
|
|
|
The following models were included in the merge: |
|
|
* [knifeayumu/Cydonia-v1.2-Magnum-v4-22B](https://huggingface.co/knifeayumu/Cydonia-v1.2-Magnum-v4-22B) |
|
|
* [Steelskull/MSM-MS-Cydrion-22B](https://huggingface.co/Steelskull/MSM-MS-Cydrion-22B) |
|
|
|
|
|
### Configuration |
|
|
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
|
|
```yaml |
|
|
models: |
|
|
- model: knifeayumu/Cydonia-v1.2-Magnum-v4-22B |
|
|
- model: Steelskull/MSM-MS-Cydrion-22B |
|
|
merge_method: slerp |
|
|
base_model: knifeayumu/Cydonia-v1.2-Magnum-v4-22B |
|
|
parameters: |
|
|
t: [0.1, 0.2, 0.4, 0.6, 0.6, 0.4, 0.2, 0.1] |
|
|
dtype: bfloat16 |
|
|
tokenizer_source: base |
|
|
|
|
|
``` |
|
|
|
|
|
</p> |
|
|
</details> |
|
|
|
|
|
|
