johnsutor/mixture-of-llamas-linear

linear

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

Merge Details

Merge Method

This model was merged using the linear merge method using meta-llama/Meta-Llama-3-8B-Instruct as a base.

Models Merged

The following models were included in the merge:

• VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct
• nbeerbower/llama-3-gutenberg-8B
• jpacifico/French-Alpaca-Llama3-8B-Instruct-v1.0
• DeepMount00/Llama-3-8b-Ita
• failspy/Meta-Llama-3-8B-Instruct-abliterated-v3

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: meta-llama/Meta-Llama-3-8B-Instruct
    parameters:
      density: 0.5
      weight: 1.0
  - model: failspy/Meta-Llama-3-8B-Instruct-abliterated-v3
    parameters:
      density: 0.5
      weight: 1.0
  - model: VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct
    parameters:
      density: 0.5
      weight: 1.0
  - model: DeepMount00/Llama-3-8b-Ita
    parameters:
      density: 0.5
      weight: 1.0
  - model: nbeerbower/llama-3-gutenberg-8B
    parameters:
      density: 0.5
      weight: 1.0
  - model: jpacifico/French-Alpaca-Llama3-8B-Instruct-v1.0
    parameters:
      density: 0.5
      weight: 1.0
merge_method: linear
tokenizer_source: union
base_model: meta-llama/Meta-Llama-3-8B-Instruct
parameters:
  int8_mask: true
dtype: bfloat16