merged

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

Model merge focus was for uncensored Role Playing.
Please be aware that this model is for fictional and entertainment purposes.
Any other sort of usage is out of scope.

Merge Details

Merge Method

This model was merged using the Model Stock merge method using models/Chronos-Gold-12B-1.0 as a base.

Models Merged

The following models were included in the merge:

P0x0/Astra-v1-12B
elinas/Chronos-Gold-12B-1.0
dphn/dolphin-2.9.3-mistral-nemo-12b
LatitudeGames/Muse-12B
Gryphe/Pantheon-RP-1.6-12b-Nemo
PygmalionAI/Pygmalion-3-12B
LatitudeGames/Wayfarer-2-12B
aixonlab/Zinakha-12b

Configuration

The following YAML configuration was used to produce this model:

base_model: models/Chronos-Gold-12B-1.0
dtype: bfloat16
merge_method: model_stock
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 40]
        model: P0x0/Astra-v1-12B
      - layer_range: [0, 40]
        model: elinas/Chronos-Gold-12B-1.0
      - layer_range: [0, 40]
        model: dphn/dolphin-2.9.3-mistral-nemo-12b
      - layer_range: [0, 40]
        model: LatitudeGames/Muse-12B
      - layer_range: [0, 40]
        model: Gryphe/Pantheon-RP-1.6-12b-Nemo
      - layer_range: [0, 40]
        model: PygmalionAI/Pygmalion-3-12B
      - layer_range: [0, 40]
        model: LatitudeGames/Wayfarer-2-12B
      - layer_range: [0, 40]
        model: aixonlab/Zinakha-12b
parameters:
  normalize: 1.0
tokenizer_source: LatitudeGames/Muse-12B