AlexCuadron/dpo_plroleplay

output

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

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• Nitral-AI/Captain-Eris_Violet-GRPO-v0.420
• AlexCuadron/dpo_roleplay

Configuration

The following YAML configuration was used to produce this model:

merge_method: slerp
slices:
  - sources:
      - model: Nitral-AI/Captain-Eris_Violet-GRPO-v0.420
        layer_range: [0, 40]
      - model: AlexCuadron/dpo_roleplay
        layer_range: [0, 40]
    base_model: Nitral-AI/Captain-Eris_Violet-GRPO-v0.420
    parameters:
      t:
        - filter: self_attn
          value: [0.6, 0.5, 0.5, 0.4, 0.3]
        - filter: mlp
          value: [0.4, 0.5, 0.5, 0.7, 0.8]
        - filter: "layer=[0:10]"
          value: 0.5
        - filter: "layer=[30:40]"
          value: 0.6
        - value: 0.65
    dtype: bfloat16