MrRobotoAI/Hathor-v4.2

merge

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

Merge Details

Merge Method

This model was merged using the linear DARE merge method using MrRobotoAI/Hathor-v3.5 as a base.

Models Merged

The following models were included in the merge:

• NousResearch/Yarn-Mistral-7b-128k
• Norquinal/Mistral-7B-storywriter
• Aratako/SniffyOtter-7B-Novel-Writing-NSFW
• OmnicromsBrain/Eros_Scribe-7b
• CallComply/zephyr-7b-beta-128k
• Aratako/Antler-7B-Novel-Writing
• Nitral-Archive/HerculeanSea-7b-128k

Configuration

The following YAML configuration was used to produce this model:

 models:

  - model: CallComply/zephyr-7b-beta-128k
    parameters:
      weight: 0.125
      density: 0.9
  - model: Nitral-Archive/HerculeanSea-7b-128k
    parameters:
      weight: 0.125
      density: 0.9
  - model: NousResearch/Yarn-Mistral-7b-128k
    parameters:
      weight: 0.125
      density: 0.9
  - model: Norquinal/Mistral-7B-storywriter
    parameters:
      weight: 0.0625
      density: 0.9
  - model: MrRobotoAI/Hathor-v3.5
    parameters:
      weight: 0.3125
      density: 0.9
  - model: OmnicromsBrain/Eros_Scribe-7b
    parameters:
      weight: 0.125
      density: 0.9
  - model: Aratako/Antler-7B-Novel-Writing
    parameters:
      weight: 0.0625
      density: 0.9
  - model: Aratako/SniffyOtter-7B-Novel-Writing-NSFW
    parameters:
      weight: 0.0625
      density: 0.9

 merge_method: dare_linear
 base_model: MrRobotoAI/Hathor-v3.5
 parameters:
  normalize: true
 dtype: float16