BimatrixTemp/Llama3.1-8B-SFT-TIES

Llama3.1-dare-ties

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

Merge Details

Merge Method

This model was merged using the TIES merge method using meta-llama/Llama-3.1-8B-Instruct as a base.

Models Merged

The following models were included in the merge:

• BimatrixTemp/BIMATRIX-Llama-3.1-Lora
• /workspace/mssong/finetune/mergekit/models/fcc-llama-8b

Configuration

The following YAML configuration was used to produce this model:

base_model: meta-llama/Llama-3.1-8B-Instruct
dtype: bfloat16
merge_method: ties
parameters:
  int8_mask: 1.0
  normalize: 0.0
slices:
- sources:
  - layer_range: [0, 32]
    model: BimatrixTemp/BIMATRIX-Llama-3.1-Lora
    parameters:
      density: 0.5
      weight: 0.5
  - layer_range: [0, 32]
    model: /workspace/mssong/finetune/mergekit/models/fcc-llama-8b
    parameters:
      density: 0.5
      weight: 0.5
  - layer_range: [0, 32]
    model: meta-llama/Llama-3.1-8B-Instruct