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: jpacifico/Chocolatine-3B-Instruct-DPO-Revised
5 layer_range: [0, 32]
6 - model: microsoft/Phi-3-mini-4k-instruct
7 layer_range: [0, 32]
8merge_method: slerp
9base_model: jpacifico/Chocolatine-3B-Instruct-DPO-Revised
10parameters:
11 t:
12 - filter: self_attn
13 value: [0, 0.25, 0.5, 0.75, 1]
14 - filter: mlp
15 value: [1, 0.75, 0.5, 0.25, 0]
16 - value: 0.5
17dtype: bfloat16
18
