Modelpublic

Alelcv27/Llama3.1-8B-Breadcrumbs-Ties-v1

sourceHugging Faceupdated 8mo agoView on Hugging Face

0likes12downloads

Model Card

Llama3.1-8B-Breadcrumbs-Ties-v1

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

Merge Details

Merge Method

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

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

yaml

base_model: meta-llama/Llama-3.1-8B-Instruct
dtype: float16
merge_method: breadcrumbs_ties
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 32]
        model: Alelcv27/Llama3.1-8B-Math-CoT
        parameters:
          density: 0.5
          weight: 0.8
      - layer_range: [0, 32]
        model: Alelcv27/Llama3.1-8B-Code
        parameters:
          density: 0.6
          weight: 0.9
      - layer_range: [0, 32]
        model: meta-llama/Llama-3.1-8B-Instruct
parameters:
  normalize: 1.0