This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
DARE TIES merge method using
cognitivecomputations/dolphin-2.9-llama3-8b as a base.
1models:
2 - model: cognitivecomputations/dolphin-2.9-llama3-8b
3 # no parameters necessary for base model
4 - model: Weyaxi/Einstein-v6.1-Llama3-8B
5 parameters:
6 weight: 0.5
7 density: 0.8
8merge_method: dare_ties
9base_model: cognitivecomputations/dolphin-2.9-llama3-8b
10parameters:
11 int8_mask: true
12dtype: bfloat16
