MS-MagpantheonselRP-22B-14-Test

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 unsloth/Mistral-Small-Instruct-2409 as a base.

Models Merged

The following models were included in the merge:

• Envoid/Mistral-Small-NovusKyver
• Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V1-22B
• Kaoeiri/MS_dampf-2409-22B
• ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
• Kaoeiri/MS-Inky-2409-22B
• Kaoeiri/MS_fujin-2409-22B
• DigitalSouls/BlackSheep-DigitalSoul-22B
• invisietch/MiS-Firefly-v0.2-22B
• Kaoeiri/MS_Moingooistral-2409-22B
• crestf411/MS-sunfall-v0.7.0
• Kaoeiri/Magnum-v4-Cydonia-vXXX-22B
• InferenceIllusionist/SorcererLM-22B
• qingy2024/MwM-22B-Instruct
• spow12/ChatWaifu_v2.0_22B
• Darkknight535/MS-Moonlight-22B-v3
• Kaoeiri/MS_a-coolyte-2409-22B
• Gryphe/Pantheon-RP-1.6.2-22b-Small
• allura-org/MS-Meadowlark-22B
• concedo/Beepo-22B

Configuration

The following YAML configuration was used to produce this model:

models:
  # Core Fiction and Character Detail Models
  - model: Kaoeiri/MS_Moingooistral-2409-22B
    parameters:
      weight: 0.10
      density: 0.72

  - model: Kaoeiri/Magnum-v4-Cydonia-vXXX-22B
    parameters:
      weight: 0.72
      density: 0.80

  # World Building & Character Interaction
  - model: Kaoeiri/MS-Inky-2409-22B
    parameters:
      weight: 0.24
      density: 0.70

  - model: Gryphe/Pantheon-RP-1.6.2-22b-Small
    parameters:
      weight: 0.32
      density: 0.76

  # Character Development Core
  - model: DigitalSouls/BlackSheep-DigitalSoul-22B
    parameters:
      weight: 0.12
      density: 0.68

  # Magical Elements
  - model: InferenceIllusionist/SorcererLM-22B
    parameters:
      weight: 0.10
      density: 0.70

  - model: Envoid/Mistral-Small-NovusKyver
    parameters:
      weight: 0.08
      density: 0.70

  # Secondary and Creativity Character Enhancement
  - model: concedo/Beepo-22B
    parameters:
      weight: 0.35
      density: 0.78
  
  - model: crestf411/MS-sunfall-v0.7.0
    parameters:
      weight: 0.14
      density: 0.68

  - model: Kaoeiri/MS_a-coolyte-2409-22B
    parameters:
      weight: 0.10
      density: 0.65

  - model: invisietch/MiS-Firefly-v0.2-22B
    parameters:
      weight: 0.12
      density: 0.68

  # Enhanced Personality Dynamics
  - model: Kaoeiri/MS_fujin-2409-22B
    parameters:
      weight: 0.08
      density: 0.62
  
  - model: Kaoeiri/MS_dampf-2409-22B
    parameters:
      weight: 0.10
      density: 0.62

  # Enhanced Story and Character Building
  - model: qingy2024/MwM-22B-Instruct
    parameters:
      weight: 0.18
      density: 0.72

  - model: ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
    parameters:
      weight: 0.12
      density: 0.65
  
  - model: Darkknight535/MS-Moonlight-22B-v3
    parameters:
      weight: 0.10
      density: 0.65

  - model: allura-org/MS-Meadowlark-22B
    parameters:
      weight: 0.16
      density: 0.70

  # Cultural and Character Depth
  - model: Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V1-22B
    parameters:
      weight: 0.16
      density: 0.62

  - model: spow12/ChatWaifu_v2.0_22B
    parameters:
      weight: 0.20
      density: 0.65

merge_method: dare_ties
base_model: unsloth/Mistral-Small-Instruct-2409
parameters:
  density: 0.95
  epsilon: 0.12 # I'm gonna crash out if this doesn't work properly, istg.
  lambda: 1.22
dtype: bfloat16