MrRobotoAI/Mpasila-Viking-8b

merge

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

Merge Details

Merge Method

This model was merged using the TIES merge method using MrRobotoAI/MrRoboto-BASE-v3-8b-64k as a base.

Models Merged

The following models were included in the merge:

• mpasila/OpenCrystal-8B-L3
• mpasila/Llama-3-Umbral-Mind-Instruct-8B
• mpasila/Llama-3-Umbral-Mind-Replete-8B
• mpasila/Llama-3.1-Literotica-8B
• mpasila/Llama-3-Umbral-Mind-SimPO-8B
• nothingiisreal/L3-8B-Stheno-Horny-v3.3-32K

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: MrRobotoAI/MrRoboto-BASE-v3-8b-64k
  - model: mpasila/OpenCrystal-8B-L3
    parameters:
      density: 0.1
      weight: 0.1
  - model: mpasila/Llama-3.1-Literotica-8B
    parameters:
      density: 0.2
      weight: 0.2
  - model: mpasila/Llama-3-Umbral-Mind-Replete-8B
    parameters:
      density: 0.15
      weight: 0.15
  - model: mpasila/Llama-3-Umbral-Mind-Instruct-8B
    parameters:
      density: 0.15
      weight: 0.15
  - model: mpasila/Llama-3-Umbral-Mind-SimPO-8B
    parameters:
      density: 0.15
      weight: 0.2
  - model: nothingiisreal/L3-8B-Stheno-Horny-v3.3-32K
    parameters:
      density: 0.3
      weight: 0.3
merge_method: ties
base_model: MrRobotoAI/MrRoboto-BASE-v3-8b-64k
parameters:
  int8_mask: true
  rescale: true
  normalize: false
dtype: bfloat16