This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
TIES merge method using
mistralai/Mistral-7B-v0.1 as a base.
1base_model: mistralai/Mistral-7B-v0.1
2dtype: float16
3merge_method: ties
4parameters:
5 int8_mask: 1.0
6 normalize: 1.0
7slices:
8- sources:
9 - layer_range: [0, 32]
10 model: oz1115/Marcoro14-7B-slerp_v2
11 parameters:
12 density: [1.0, 0.7, 0.1]
13 weight: 1.0
14 - layer_range: [0, 32]
15 model: mlabonne/NeuralHermes-2.5-Mistral-7B
16 parameters:
17 density: 0.33
18 weight:
19 - filter: mlp
20 value: 0.5
21 - value: 0.0
22 - layer_range: [0, 32]
23 model: mistralai/Mistral-7B-v0.1
