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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using /fred/oz446/Finetuning/model/medalpaca-7b as a base.

Models Merged

The following models were included in the merge:

  • /fred/oz446/Finetuning/model/medalpaca-kd
  • /fred/oz446/Finetuning/model/medalpaca-sft

Configuration

The following YAML configuration was used to produce this model:

base_model: /fred/oz446/Finetuning/model/medalpaca-7b
dtype: bfloat16
merge_method: dare_ties
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 32]
        model: /fred/oz446/Finetuning/model/medalpaca-sft
        parameters:
          density: 0.6
          weight: 0.2
      - layer_range: [0, 32]
        model: /fred/oz446/Finetuning/model/medalpaca-kd
        parameters:
          density: 0.6
          weight: 0.8
      - layer_range: [0, 32]
        model: /fred/oz446/Finetuning/model/medalpaca-7b
Downloads last month
6
Safetensors
Model size
7B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collection including MedSwin/MedSwin-Dare_Ties-KD-0.75-0.7

Paper for MedSwin/MedSwin-Dare_Ties-KD-0.75-0.7