RP-SAINEMO - "Попытка слиять хорошие RP модели с Ру-моделями на базе NeMo"

GGUF https://huggingface.co./mradermacher/RP-SAINEMO-GGUF

Настройки

https://huggingface.co./MarinaraSpaghetti/SillyTavern-Settings/tree/main/Customized/Mistral%20Improved Оптимально для данной модели

temp - 1,2
TopA - 0,1
ToP - 1
DRY - 0,8 1,75 2 0

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the della_linear merge method using E:\Programs\TextGen\text-generation-webui\models\IlyaGusev_saiga_nemo_12b as a base.

Models Merged

The following models were included in the merge:

• E:\Programs\TextGen\text-generation-webui\models\MarinaraSpaghetti_NemoMix-Unleashed-12B

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: E:\Programs\TextGen\text-generation-webui\models\IlyaGusev_saiga_nemo_12b
    parameters:
      weight: 0.8    # Приоритет русскому контексту
      density: 0.4
  - model: E:\Programs\TextGen\text-generation-webui\models\MarinaraSpaghetti_NemoMix-Unleashed-12B
    parameters:
      weight: 0.2    # Меньший вес, чтобы сохранить рп элементы, но не забыть про русский
      density: 0.4

merge_method: della_linear
base_model: E:\Programs\TextGen\text-generation-webui\models\IlyaGusev_saiga_nemo_12b
parameters:
  epsilon: 0.05
  lambda: 1
dtype: float16
tokenizer_source: base