Opulus-12B-v2

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using mistralai/Mistral-Nemo-Instruct-2407 as a base.

Models Merged

The following models were included in the merge:

• Entropicengine/Pinecone-Rune-12b
• EpistemeAI/Mistral-Nemo-Instruct-12B-Philosophy-Math
• ReadyArt/Forgotten-Safeword-12B-v4.0

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: mistralai/Mistral-Nemo-Instruct-2407
    parameters:
      weight: 0.40
  - model: EpistemeAI/Mistral-Nemo-Instruct-12B-Philosophy-Math
    parameters:
      weight: 0.25
  - model: Entropicengine/Pinecone-Rune-12b  
    parameters:
      weight: 0.20
  - model: ReadyArt/Forgotten-Safeword-12B-v4.0
    parameters:
      weight: 0.15

merge_method: ties
base_model: mistralai/Mistral-Nemo-Instruct-2407
dtype: float16

This is the merge made to then run the Opus (and some Sonnet 3.5) datasets on. Haven't tested it. Babsie/Opulus-12B-v3 is made from this merge.