~ The power of Three

Trifecta-Max-24b

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

Recommended ST preset for RP :

• Sphiratrioth

☕ Support My Work

If you like my work, consider buying me a coffee to support future merges, GPU time, and experiments.

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using darkc0de/XortronCriminalComputingConfig as a base.

Models Merged

The following models were included in the merge:

• TheDrummer/Cydonia-24B-v3
• Sorawiz/MistralCreative-24B-Chat

Configuration

The following YAML configuration was used to produce this model:

base_model: darkc0de/XortronCriminalComputingConfig
chat_template: auto
merge_method: dare_ties
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 40]
        model: darkc0de/XortronCriminalComputingConfig
        parameters:
          weight: 0.4
      - layer_range: [0, 40]
        model: Sorawiz/MistralCreative-24B-Chat
        parameters:
          weight: 0.3
      - layer_range: [0, 40]
        model: TheDrummer/Cydonia-24B-v3
        parameters:
          weight: 0.3
out_dtype: bfloat16
parameters:
  density: 1.0
tokenizer: {}