G4-WanaMeroQueen-v3-31B
This is a merge of pre-trained language models created using mergekit. 前回のマージvulture3/G4-MeroWana-31Bでは、やりすぎで日本語の能力が明らかに落ちていました。 日本語については必要以上に劣化しない、さらには話の筋道が通ったものとなる様、微妙な加減でのマージを目指してみました。
Merge Details
Merge Method
This model was merged using the DARE TIES merge method using kawaimasa/Wanabi-Gemma4-31B as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
merge_method: dare_ties
base_model: kawaimasa/Wanabi-Gemma4-31B
parameters:
lambda: 1.0
dtype: bfloat16
models:
- model: aifeifei798/Gemma-4-Queen-31B-it
parameters:
weight: [0.20, 0.35, 0.40, 0.05, 0.00]
density: [0.25, 0.35, 0.40, 0.05, 0.00]
- model: zerofata/G4-MeroMero-v2-31B
parameters:
weight: [0.00, 0.10, 0.25, 0.25, 0.20]
density: [0.00, 0.10, 0.25, 0.25, 0.20]