Calliope-7b

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

Merge Details

Merge Method

This model was merged using the rescaled_sample merge method using mistralai/Mistral-7B-v0.1 as a base.

Models Merged

The following models were included in the merge:

• Eric111/openchat-3.5-0106-128k-DPO_dpo-binarized-NeuralTrix-7B
• Nexusflow/Starling-LM-7B-beta
• paulml/OmniBeagleSquaredMBX-v3-7B
• chargoddard/servile-harpsichord-cdpo

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Nexusflow/Starling-LM-7B-beta
    parameters:
      weight: 0.5
      density: 0.5
  - model: Eric111/openchat-3.5-0106-128k-DPO_dpo-binarized-NeuralTrix-7B
    parameters:
      weight: 0.3
      density: 0.3
  - model: paulml/OmniBeagleSquaredMBX-v3-7B
    parameters:
      weight: 0.2
      density: 0.2
  - model: chargoddard/servile-harpsichord-cdpo
    parameters:
      weight: 0.3
      density: 0.3
merge_method: rescaled_sample
base_model: mistralai/Mistral-7B-v0.1
parameters:
  int8_mask: true
dtype: bfloat16