This is probably the first multislerp model on Hugging Face.

Emilia-Multislerp-12B

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

Merge Details

Merge Method

This model was merged using the Multi-SLERP merge method using yamatazen/Orihime-12B as a base.

Models Merged

The following models were included in the merge:

• natong19/Mistral-Nemo-Instruct-2407-abliterated
• nbeerbower/mistral-nemo-gutenberg-12B-v4

Configuration

The following YAML configuration was used to produce this model:

base_model: yamatazen/Orihime-12B
models:
  - model: nbeerbower/mistral-nemo-gutenberg-12B-v4
    parameters:
      weight: [0.25, 0.3, 0.5, 0.6, 0.75]
  - model: natong19/Mistral-Nemo-Instruct-2407-abliterated
    parameters:
      weight: [0.25, 0.3, 0.5, 0.3, 0.25]
merge_method: multislerp
dtype: bfloat16
out_dtype: bfloat16
parameters:
  normalize: true
tokenizer:
  source: union