README.md

metadata

license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - ValiantLabs/Llama3.1-8B-Cobalt

ZeroXClem/Llama3.1-8B-Titanium-Forge

ZeroXClem/Llama3.1-8B-Titanium-Forge is a merge of the following models using mergekit:

• ValiantLabs/Llama3.1-8B-Cobalt

🧩 Configuration

base_model: ValiantLabs/Llama3.1-8B-ShiningValiant2
dtype: float32
merge_method: model_stock
slices:
  - sources:
    - layer_range: [0, 32]
      model: ValiantLabs/Llama3.1-8B-Cobalt
    - layer_range: [0, 32]
      model: ValiantLabs/Llama3.1-8B-Fireplace2
    - layer_range: [0, 32]
      model: ValiantLabs/Llama3.1-8B-Enigma
    - layer_range: [0, 32]
      model: ValiantLabs/Llama3.1-8B-ShiningValiant2
out_dtype: bfloat16