Supposedly halfway between
knifeayumu/Cydonia-v1.2-Magnum-v4-22B and
knifeayumu/Cydonia-v1.3-Magnum-v4-22B. For some reason, I used a different method instead of merging both of those... IDK why but if it works well, give me a thumbs up(?)
This is a merge of pre-trained language models created using
mergekit.
This model was merged using the SLERP merge method.
1models:
2 - model: knifeayumu/Cydonia-v1.2-v1.3-22B
3 - model: anthracite-org/magnum-v4-22b
4merge_method: slerp
5base_model: knifeayumu/Cydonia-v1.2-v1.3-22B
6parameters:
7 t: [0.1, 0.3, 0.6, 0.3, 0.1]
8dtype: bfloat16