Qwen-3B-Instruct-Karcher-v1

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

Merge Details

Merge Method

This model was merged using the Karcher Mean merge method using Qwen/Qwen2.5-3B-Instruct as a base.

Models Merged

The following models were included in the merge:

• Alelcv27/Qwen2.5-3B-Instruct-Code
• Alelcv27/Qwen2.5-3B-Instruct-Math-CoT

Configuration

The following YAML configuration was used to produce this model:

base_model: Qwen/Qwen2.5-3B-Instruct
dtype: bfloat16
merge_method: karcher
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 36]
        model: Alelcv27/Qwen2.5-3B-Instruct-Math-CoT
      - layer_range: [0, 36]
        model: Alelcv27/Qwen2.5-3B-Instruct-Code
      - layer_range: [0, 36]
        model: Qwen/Qwen2.5-3B-Instruct
parameters:
  max_iter: 100.0
tokenizer_source: base