|
--- |
|
base_model: |
|
- win10/Mistral-Nemo-abliterated-Nemo-Pro-v2 |
|
- PocketDoc/Dans-PersonalityEngine-V1.1.0-12b |
|
- Sao10K/MN-BackyardAI-Party-12B-v1 |
|
- IlyaGusev/saiga_nemo_12b |
|
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 [win10/Mistral-Nemo-abliterated-Nemo-Pro-v2](https://huggingface.co/win10/Mistral-Nemo-abliterated-Nemo-Pro-v2) as a base. |
|
|
|
### Models Merged |
|
|
|
The following models were included in the merge: |
|
* [PocketDoc/Dans-PersonalityEngine-V1.1.0-12b](https://huggingface.co/PocketDoc/Dans-PersonalityEngine-V1.1.0-12b) |
|
* [Sao10K/MN-BackyardAI-Party-12B-v1](https://huggingface.co/Sao10K/MN-BackyardAI-Party-12B-v1) |
|
* [IlyaGusev/saiga_nemo_12b](https://huggingface.co/IlyaGusev/saiga_nemo_12b) |
|
|
|
### Configuration |
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
```yaml |
|
models: |
|
- model: PocketDoc/Dans-PersonalityEngine-V1.1.0-12b |
|
parameters: |
|
density: 1 |
|
weight: 1 |
|
|
|
- model: Sao10K/MN-BackyardAI-Party-12B-v1 |
|
parameters: |
|
density: 1 |
|
weight: 1 |
|
|
|
- model: IlyaGusev/saiga_nemo_12b |
|
parameters: |
|
density: 1 |
|
weight: 1 |
|
|
|
merge_method: ties |
|
base_model: win10/Mistral-Nemo-abliterated-Nemo-Pro-v2 |
|
|
|
dtype: float32 |
|
chat_template: chatml |
|
# Regularization |
|
regularization: |
|
- method: gradient_penalty |
|
scale: 0.05 # Increased influence for gradient control |
|
- method: weight_clipping |
|
clip_range: [-0.2, 0.2] # Broader clipping range for flexibility |
|
- method: random_noise |
|
scale: 0.01 # Stronger noise injection |
|
- method: attention_dropout |
|
scale: 0.1 # Higher dropout to reduce attention fixation |
|
|
|
# Postprocessing |
|
postprocessing: |
|
- operation: entropy_regularization |
|
scale: 0.05 # Stronger encouragement for diverse outputs |
|
- operation: non_linear_scaling |
|
parameters: |
|
function: tanh |
|
- operation: sharpening |
|
intensity: 0.5 # Enhanced sharpening for precise outputs |
|
- operation: gaussian_smoothing |
|
sigma: 1.5 # Increased smoothing for stable outputs |
|
- operation: normalize |
|
- operation: dynamic_scaling |
|
scale_range: [0.8, 1.2] # Expanded dynamic range for scaling |
|
- operation: smoothing |
|
parameters: |
|
adaptive: true |
|
range: [0.85, 1.15] # Wider adaptive smoothing range |
|
kernel_size: 5 |
|
|
|
``` |
|
|
