Modelpublic

Alignment-Lab-AI/Mistral-nemo-3b-unhealed

sourceHugging Faceupdated 2y agoView on Hugging Face

3likes199downloads

Model Card

output

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

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

—ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.3
—mistralai/Mistral-Nemo-Base-2407

Configuration

The following YAML configuration was used to produce this model:

yaml

dtype: bfloat16
merge_method: passthrough
slices:
  - sources:
    - layer_range: [0, 7]  # Taking first 13 layers (0-12)
      model: ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.3
  - sources:
    - layer_range: [18, 19]  # Taking last 2 layers for output stability
      model: mistralai/Mistral-Nemo-Base-2407
  - sources:
    - layer_range: [32, 39]  # Taking last 2 layers for output stability
      model: mistralai/Mistral-Nemo-Base-2407