kromcomp/L3.1-Shakev3-10B

shake

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/slurp

Configuration

The following YAML configuration was used to produce this model:

dtype: float32
merge_method: passthrough
parameters:
  int8_mask: 1.0
slices:
- sources:
  - layer_range: [0, 20]
    model: merge/slurp
- sources:
  - layer_range: [16, 20]
    model: merge/slurp
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
- sources:
  - layer_range: [20, 28]
    model: merge/slurp
- sources:
  - layer_range: [24, 30]
    model: merge/slurp
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
- sources:
  - layer_range: [28, 32]
    model: merge/slurp