EvoMed

final_model

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 lightblue/suzume-llama-3-8B-japanese as a base.

Models Merged

The following models were included in the merge:

• ContactDoctor/Bio-Medical-Llama-3-8B

Configuration

The following YAML configuration was used to produce this model:

base_model: lightblue/suzume-llama-3-8B-japanese
dtype: bfloat16
merge_method: dare_ties
parameters:
  int8_mask: 1.0
  normalize: 1.0
slices:
- sources:
  - layer_range: [0, 8]
    model: ContactDoctor/Bio-Medical-Llama-3-8B
    parameters:
      density: 1.0
      weight: 0.6133219045070127
  - layer_range: [0, 8]
    model: lightblue/suzume-llama-3-8B-japanese
    parameters:
      density: 0.685860266951033
      weight: 0.5895381594604311
- sources:
  - layer_range: [8, 16]
    model: ContactDoctor/Bio-Medical-Llama-3-8B
    parameters:
      density: 0.7392837955301343
      weight: 0.3228829047267915
  - layer_range: [8, 16]
    model: lightblue/suzume-llama-3-8B-japanese
    parameters:
      density: 1.0
      weight: 0.6225596018347737
- sources:
  - layer_range: [16, 24]
    model: ContactDoctor/Bio-Medical-Llama-3-8B
    parameters:
      density: 1.0
      weight: 0.6675711396324198
  - layer_range: [16, 24]
    model: lightblue/suzume-llama-3-8B-japanese
    parameters:
      density: 1.0
      weight: 0.507981935427293
- sources:
  - layer_range: [24, 32]
    model: ContactDoctor/Bio-Medical-Llama-3-8B
    parameters:
      density: 0.7479105312794881
      weight: 0.6307368863287528
  - layer_range: [24, 32]
    model: lightblue/suzume-llama-3-8B-japanese
    parameters:
      density: 0.7322891014425874
      weight: 0.633799814811044