Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Paper
•
2311.03099
•
Published
•
30
This is a merge of pre-trained language models created using mergekit.
This model was merged using the linear DARE merge method using llm-jp/llm-jp-3-13b as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: llm-jp/llm-jp-3-13b
- model: llm-jp/llm-jp-3-13b-instruct
parameters:
weight: 1
merge_method: dare_linear
base_model: llm-jp/llm-jp-3-13b
parameters:
normalize: true
dtype: bfloat16