This is a merge of pre-trained language models created using
mergekit.
This model was merged using the passthrough merge method.
1slices:
2- sources:
3 - layer_range: [0, 3]
4 model: Qwen/Qwen2.5-0.5B-Instruct
5- sources:
6 - layer_range: [4, 7]
7 model: Qwen/Qwen2.5-0.5B-Instruct
8- sources:
9 - layer_range: [8, 11]
10 model: Qwen/Qwen2.5-0.5B-Instruct
11- sources:
12 - layer_range: [12, 15]
13 model: Qwen/Qwen2.5-0.5B-Instruct
14- sources:
15 - layer_range: [16, 19]
16 model: Qwen/Qwen2.5-0.5B-Instruct
17- sources:
18 - layer_range: [20, 23]
19 model: Qwen/Qwen2.5-0.5B-Instruct
20merge_method: passthrough
21dtype: float16
22
