This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
DARE TIES merge method using
ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1 as a base.
1merge_method: dare_ties
2base_model: ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
3parameters:
4 int8_mask: true
5dtype: bfloat16
6models:
7 - model: ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
8 parameters:
9 weight: 0.30
10 density: 0.78
11 - model: anthracite-org/magnum-v4-22b
12 parameters:
13 weight: 0.25
14 density: 0.66
15 - model: allura-org/MS-Meadowlark-22B
16 parameters:
17 weight: 0.20
18 density: 0.54
19 - model: rAIfle/Acolyte-22B
20 parameters:
21 weight: 0.15
22 density: 0.42
23 - model: Gryphe/Pantheon-RP-1.6.2-22b-Small
24 parameters:
25 weight: 0.10
26 density: 0.42
