mergekit-community/MN-Sappho-m-12B

merge

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

Merge Details

Merge Method

This model was merged using the Model Stock merge method using Khetterman/AbominationScience-12B-v4 as a base.

Models Merged

The following models were included in the merge:

• mergekit-community/MN-Sappho-g2-12B
• mergekit-community/MN-Sappho-e-12B
• mergekit-community/MN-Sappho-j-12B
• mergekit-community/MN-Sappho-k-12B + jtatman/mistral_nemo_12b_reasoning_psychology_lora
• mistralai/Mistral-Nemo-Instruct-2407 + XeroCodes/aurora-12b
• LatitudeGames/Wayfarer-12B + jtatman/mistral_nemo_12b_reasoning_psychology_lora
• mistralai/Mistral-Nemo-Base-2407

Configuration

The following YAML configuration was used to produce this model:

dtype: float32
out_dtype: bfloat16
merge_method: model_stock
base_model: Khetterman/AbominationScience-12B-v4
models:
  - model: mergekit-community/MN-Sappho-k-12B+jtatman/mistral_nemo_12b_reasoning_psychology_lora
    parameters:
      weight: 0.5
  - model: mergekit-community/MN-Sappho-j-12B
    parameters:
      weight: 1.3
  - model: mergekit-community/MN-Sappho-g2-12B
    parameters:
      weight: 1.3
  - model: mergekit-community/MN-Sappho-e-12B
  - model: LatitudeGames/Wayfarer-12B+jtatman/mistral_nemo_12b_reasoning_psychology_lora
  - model: mistralai/Mistral-Nemo-Instruct-2407+XeroCodes/aurora-12b
  - model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      weight: 1.1