kromcomp/L3.1-Milkywayv1-12B

milkyway

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

Merge Details

Merge Method

This model was merged using the Passthrough merge method.

Models Merged

The following models were included in the merge:

• merge/stars

Configuration

The following YAML configuration was used to produce this model:

chat_template: llama3
dtype: float32
merge_method: passthrough
slices:
- sources:
  - layer_range: [0, 8]
    model: merge/stars
- sources:
  - layer_range: [4, 8]
    model: merge/stars
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
- sources:
  - layer_range: [8, 12]
    model: merge/stars
- sources:
  - layer_range: [8, 12]
    model: merge/stars
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
- sources:
  - layer_range: [12, 16]
    model: merge/stars
- sources:
  - layer_range: [12, 16]
    model: merge/stars
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
- sources:
  - layer_range: [16, 28]
    model: merge/stars
- sources:
  - layer_range: [24, 28]
    model: merge/stars
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
- sources:
  - layer_range: [28, 32]
    model: merge/stars
- sources:
  - layer_range: [30, 32]
    model: merge/stars
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0