This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
DARE TIES merge method using
unsloth/Llama-3.2-3B-Instruct as a base.
1models:
2 - model: unsloth/Llama-3.2-3B-Instruct
3 # No parameters necessary for base model
4 - model: aabberick/Llama3.2-3b-H200-End_16bit
5 parameters:
6 density: 0.9
7 weight: 0.5
8 - model: aabberick/Llama3.2-3b-H200FamilyM
9 parameters:
10 density: 0.9
11 weight: 0.5
12merge_method: dare_ties
13base_model: unsloth/Llama-3.2-3B-Instruct
14parameters:
15 int8_mask: false
16dtype: bfloat16
17
