Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
Paper • 2203.05482 • Published • 8
Back to basics, a simple linear merge..
In testing..
This is a merge of pre-trained language models created using mergekit.
This model was merged using the linear merge method using Undi95/PsyMedRP-v1-20B as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: Undi95/PsyMedRP-v1-20B
parameters:
density: 1
weight: 0.5
- model: Undi95/MXLewd-L2-20B
parameters:
density: 1
weight: 0.5
merge_method: linear
base_model: Undi95/PsyMedRP-v1-20B
parameters:
normalize: false
dtype: float16