README.md

metadata

base_model:
  - Xiaojian9992024/Qwen2.5-Dyanka-7B-Preview
  - Qwen/Qwen2.5-7B-Instruct
  - gz987/qwen2.5-7b-cabs-v0.3
library_name: transformers
tags:
  - mergekit
  - merge

dare_ties

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 Qwen/Qwen2.5-7B-Instruct as a base.

Models Merged

The following models were included in the merge:

• Xiaojian9992024/Qwen2.5-Dyanka-7B-Preview
• gz987/qwen2.5-7b-cabs-v0.3

Configuration

The following YAML configuration was used to produce this model:

base_model: Qwen/Qwen2.5-7B-Instruct
dtype: float16
merge_method: dare_ties
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 28]
        model: Xiaojian9992024/Qwen2.5-Dyanka-7B-Preview
        parameters:
          density: 0.8773638560987191
          weight: 1.0832441422096908
      - layer_range: [0, 28]
        model: gz987/qwen2.5-7b-cabs-v0.3
        parameters:
          density: 0.8773638560987191
          weight: 1.0832441422096908
      - layer_range: [0, 28]
        model: Qwen/Qwen2.5-7B-Instruct
parameters:
  normalize: 0.0