picAIso/code-stral-mini

merge

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:

• mistral-community/Codestral-22B-v0.1

Configuration

The following YAML configuration was used to produce this model:

  slices:
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 0
          - 8
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 7
          - 8
        parameters:
          scale:
            - filter: o_proj
              value: 0
            - filter: down_proj
              value: 0
            - value: 1
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 16
          - 24
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 23
          - 24
        parameters:
          scale:
            - filter: o_proj
              value: 0
            - filter: down_proj
              value: 0
            - value: 1
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 32
          - 40
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 39
          - 40
        parameters:
          scale:
            - filter: o_proj
              value: 0
            - filter: down_proj
              value: 0
            - value: 1
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 48
          - 56
  - sources:
      - model: mistral-community/Codestral-22B-v0.1
        layer_range:
          - 55
          - 56
        parameters:
          scale:
            - filter: o_proj
              value: 0
            - filter: down_proj
              value: 0
            - value: 1

  merge_method: passthrough
  dtype: bfloat16