Resolving Interference When Merging Models
Paper • 2306.01708 • Published • 17
This is a merge of pre-trained language models created using mergekit.
This model was merged using the TIES merge method using Guilherme34/Samantha-v6.5 as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: Guilherme34/Samantha-v6.5
# no parameters necessary for base model
- model: NousResearch/Meta-Llama-3.1-8B
parameters:
density: 0.5
weight: 0.5
merge_method: ties
base_model: Guilherme34/Samantha-v6.5
parameters:
normalize: true
dtype: float16