This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
DARE TIES merge method using
Qwen/Qwen2.5-7B-Instruct as a base.
1
2models:
3 - model: Qwen/Qwen2.5-7B-Instruct
4 parameters:
5 weight: 0.3
6 - model: Qwen/Qwen2.5-Coder-7B-Instruct
7 parameters:
8 weight: 0.4
9 - model: deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
10 parameters:
11 weight: 0.3
12
13merge_method: dare_ties
14base_model: Qwen/Qwen2.5-7B-Instruct
15parameters:
16 density: 0.8
17 normalize: true
18dtype: bfloat16
19
