This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
SLERP merge method.
1
2slices:
3 - sources:
4 - model: natong19/Mistral-Nemo-Instruct-2407-abliterated
5 layer_range: [0, 40]
6 - model: anthracite-org/magnum-v4-12b
7 layer_range: [0, 40]
8merge_method: slerp
9base_model: natong19/Mistral-Nemo-Instruct-2407-abliterated
10parameters:
11 t:
12 - filter: self_attn
13 value: [0.4, 0.6]
14 - filter: mlp
15 value: [0.3, 0.7]
16 - value: 0.5
17dtype: bfloat16
18tokenizer_source: base
19
