Kaoeiri/Magpantheon-lark-v4x1.6.2-Cydonia-vXXX-22B

merge

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using anthracite-org/magnum-v4-22b as a base.

Models Merged

The following models were included in the merge:

• TheDrummer/Cydonia-22B-v1.3
• Gryphe/Pantheon-RP-Pure-1.6.2-22b-Small
• TheDrummer/Cydonia-22B-v1.2
• allura-org/MS-Meadowlark-22B
• TheDrummer/Cydonia-22B-v1.1

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: anthracite-org/magnum-v4-22b
    parameters:
      weight: 1.0         # Magnum model for primary writing style
      density: 0.85      # Slightly lower density to allow others to contribute without overpowering
  - model: TheDrummer/Cydonia-22B-v1.3
    parameters:
      weight: 0.35        # Reduced weight for creativity to avoid too much overlap
      density: 0.75      # Reduced density for balance in creative influence
  - model: TheDrummer/Cydonia-22B-v1.2
    parameters:
      weight: 0.2         # Lower weight for extra creativity without overwhelming the story
      density: 0.7       # Lower density to reduce creative interference
  - model: TheDrummer/Cydonia-22B-v1.1
    parameters:
      weight: 0.3         # Lower weight for accuracy and specific features (evil/trolling)
      density: 0.75      # Moderate density for accuracy retention without interference
  - model: Gryphe/Pantheon-RP-Pure-1.6.2-22b-Small
    parameters:
      weight: 0.4         # Increased weight for storytelling focus
      density: 0.85      # Moderate density to allow good stories without overpowering other influences
  - model: allura-org/MS-Meadowlark-22B
    parameters:
      weight: 0.4         # Increased weight for creativity and writing balance
      density: 0.75       # Moderate density to ensure creativity enhances without overwhelming other influences
merge_method: dare_ties  # Using dare_ties for smoother blending
base_model: anthracite-org/magnum-v4-22b
parameters:
  density: 0.85          # General density for better balance between all models
  epsilon: 0.1           # Maximum change in drop probabilities
  lambda: 1.2            # Scaling factor for final merged deltas
dtype: bfloat16