This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
Linear DELLA merge method using
google/gemma-3-27b-pt as a base.
1models:
2 - model: google/gemma-3-27b-it
3 parameters:
4 weight: 0.25
5 density: 0.4
6 - model: mlabonne/gemma-3-27b-it-abliterated
7 parameters:
8 weight: 0.25
9 density: 0.2
10 - model: Lambent/mgv-27B
11 parameters:
12 weight: 0.25
13 density: 0.4
14merge_method: della_linear
15base_model: google/gemma-3-27b-pt
16parameters:
17 epsilon: 0.05
18 lambda: 1
19dtype: bfloat16
20tokenizer_source: base
21
