metadata
base_model:
- Delta-Vector/Rei-V3-KTO-12B
- grimjim/mistralai-Mistral-Nemo-Base-2407
- grimjim/AbMagnolia-v1-12B
- grimjim/mistralai-Mistral-Nemo-Instruct-2407
library_name: transformers
tags:
- mergekit
- merge
license: apache-2.0
pipeline_tag: text-generation
grimjim/Equatorium-v3-12B
This is a merge of pre-trained language models created using mergekit.
There seemed to be an instability with a prior merge component that resulted in looping, so I went for a Magnum-adjacent model instead this time. This one seems to hold up for stability and coherence under temp=1.0, minP=0.02, and the usual formatting for this class of model.
Merge Details
Merge Method
This model was merged using the Task Arithmetic merge method using grimjim/mistralai-Mistral-Nemo-Base-2407 as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
base_model: grimjim/mistralai-Mistral-Nemo-Base-2407
dtype: bfloat16
merge_method: task_arithmetic
parameters:
normalize: true
models:
- model: grimjim/mistralai-Mistral-Nemo-Base-2407
- layer_range: [0, 10]
model: grimjim/mistralai-Mistral-Nemo-Instruct-2407
parameters:
weight: 1.00
- layer_range: [10, 36]
model: grimjim/AbMagnolia-v1-12B
parameters:
weight: 0.75
- layer_range: [10, 36]
model: Delta-Vector/Rei-V3-KTO-12B
parameters:
weight: 0.25
- layer_range: [36, 40]
model: grimjim/mistralai-Mistral-Nemo-Instruct-2407
parameters:
weight: 1.00