yoshility/evo-model-merge-v1

final_merge_evo

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 ./storage_evo/input_models/Mistral-7B-v0.1_8133861 as a base.

Models Merged

The following models were included in the merge:

• ./storage_evo/input_models/WizardMath-7B-V1.1_2027605156
• ./storage_evo/input_models/shisa-gamma-7b-v1_4025154171

Configuration

The following YAML configuration was used to produce this model:

base_model: ./storage_evo/input_models/Mistral-7B-v0.1_8133861
dtype: bfloat16
merge_method: dare_ties
parameters:
  int8_mask: 1.0
  normalize: 1.0
slices:
- sources:
  - layer_range: [0, 32]
    model: ./storage_evo/input_models/shisa-gamma-7b-v1_4025154171
    parameters:
      density: 1.0
      weight: 0.011205095488974873
  - layer_range: [0, 32]
    model: ./storage_evo/input_models/WizardMath-7B-V1.1_2027605156
    parameters:
      density: 0.5651072788500664
      weight: 0.7263645106101857
  - layer_range: [0, 32]
    model: ./storage_evo/input_models/Mistral-7B-v0.1_8133861