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 DARE TIES merge method using jeiku/ToxicNoRobotsRosaHermesBoros_3B as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: jeiku/ToxicNoRobotsRosaHermesBoros_3B+jeiku/Everything_v3_StableLM
parameters:
weight: 0.25
density: 0.25
- model: jeiku/ToxicNoRobotsRosaHermesBoros_3B+jeiku/Theory_of_Mind_StableLM
parameters:
weight: 0.25
density: 0.25
merge_method: dare_ties
base_model: jeiku/ToxicNoRobotsRosaHermesBoros_3B
parameters:
dtype: bfloat16
Base model
jeiku/ToxicNoRobotsRosaHermesBoros_3B