Inspire-7B-slerp

merge

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:

• cognitivecomputations/dolphin-2.8-mistral-7b-v02
• mlabonne/AlphaMonarch-7B

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
      - model: mlabonne/AlphaMonarch-7B #mistralai/Mistral-7B-Instruct-v0.2
        layer_range: [0, 32]
      # - model: EmbeddedLLM/Mistral-7B-Merge-14-v0.1
      #   layer_range: [0, 32]
      # - model: mlabonne/AlphaMonarch-7B
      #   layer_range: [0, 32]
      - model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
        layer_range: [0, 32]
        #mlabonne/AlphaMonarch-7B

merge_method: slerp
base_model: cognitivecomputations/dolphin-2.8-mistral-7b-v02 #mistralai/Mistral-7B-Instruct-v0.2
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16