This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
TIES merge method using
unsloth/llama-2-13b as a base.
1base_model: unsloth/llama-2-13b
2chat_template: auto
3dtype: bfloat16
4merge_method: ties
5modules:
6 default:
7 slices:
8 - sources:
9 - layer_range: [0, 40]
10 model: WizardLMTeam/WizardLM-13B-V1.2
11 parameters:
12 weight: 1.0
13 - layer_range: [0, 40]
14 model: layoric/llama-2-13b-code-alpaca
15 parameters:
16 weight: 1.0
17 - layer_range: [0, 40]
18 model: unsloth/llama-2-13b
