Magrathic-12B

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

cgato/Nemo-12b-Humanize-SFT-v0.2.5-KTO was added in at very low weight in order to perturb model outputs more toward humanized text.

Tested with temperature=1.0, minP=0.02; original Nemo Instruct 2407 12B prompt format suffices.

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:

• cgato/Nemo-12b-Humanize-SFT-v0.2.5-KTO
• grimjim/Magnolia-Mell-v1-12B

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
  - model: grimjim/Magnolia-Mell-v1-12B
    parameters:
      weight: 0.997
  - model: cgato/Nemo-12b-Humanize-SFT-v0.2.5-KTO
    parameters:
      weight: 0.003