Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Paper • 2311.03099 • Published • 32
This is a merge of pre-trained language models created using mergekit.
This model was merged using the DARE TIES merge method using deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: Zachary1150/math_acc_1.5B
parameters:
weight: 0.5
density: 0.2
- model: Zachary1150/math_len_1.5B
parameters:
weight: 0.5
density: 0.2
merge_method: dare_ties
parameters:
normalize: true
lambda: 1.0
dtype: bfloat16
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B