This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
Model Stock merge method using
Qwen/Qwen2.5-7B-Instruct as a base.
1slices:
2 - sources:
3 - model: Qwen/Qwen2.5-7B-Instruct
4 layer_range: [0, 7]
5 - model: Qwen/Qwen2.5-7B-Instruct
6 layer_range: [7, 14]
7 - model: Qwen/Qwen2.5-7B-Instruct
8 layer_range: [14, 21]
9 - model: Qwen/Qwen2.5-7B-Instruct
10 layer_range: [21, 28]
11merge_method: model_stock
12base_model: Qwen/Qwen2.5-7B-Instruct
13dtype: bfloat16
