This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
TIES merge method using
OpenPipe/Qwen3-14B-Instruct as a base.
1merge_method: ties
2base_model: OpenPipe/Qwen3-14B-Instruct
3models:
4 - model: OpenPipe/Qwen3-14B-Instruct
5 parameters:
6 weight: 0.5
7 - model: Mohaddz/Qwen3-14B-Instruct-LoRA-1310-Merged
8 parameters:
9 weight: 0.3
10 - model: Qwen/Qwen3-14B-Base
11 parameters:
12 weight: 0.2
13parameters:
14 int8_mask: true
15dtype: bfloat16
16
