ping98k/deepseek-r1-distill-typhoon2-qwen2.5-7b-dare-linear-02

merge

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

Merge Details

Merge Method

This model was merged using the Linear DARE merge method using Qwen/Qwen2.5-7B as a base.

Models Merged

The following models were included in the merge:

• scb10x/typhoon2-qwen2.5-7b-instruct
• deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
• Qwen/Qwen2.5-7B-Instruct

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Qwen/Qwen2.5-7B
  - model: scb10x/typhoon2-qwen2.5-7b-instruct
    parameters:
      density: 0.80
      weight: [0.2, 0.2, 0.5, 0.8]
  - model: deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
    parameters:
      density: 0.50
      weight: [0.1, 0.5, 0.5, 0.2]
  - model: Qwen/Qwen2.5-7B-Instruct
    parameters:
      density: 0.2
      weight: [0.7, 0.3, 0.0, 0.0]
merge_method: dare_linear
base_model: Qwen/Qwen2.5-7B
parameters:
  normalize: true
dtype: bfloat16
tokenizer:
  source: scb10x/typhoon2-qwen2.5-7b-instruct